AN EXAMINATION OF ALTERNATIVE TRADING TECHNIQUES USING INTRADAY

EUR/USD CURRENCY PRICES

by

Brock Vaughters

_______________________

Doctoral Study Submitted in Partial Fulfillment

of the Requirements for the Degree of

Doctor of Business Administration

______________________

Liberty University

August 2018
Abstract

Global financial institutions provide a mechanism for multinational corporations to hedge against

exchange rate risk via currency futures contracts and spot exchange rates. Currency managers

working at these global financial institutions overseeing EUR/USD spot currency traders lack

adequate data to determine if alternative trading tools could increase net gains for their respective

firms. The purpose of this quantitative study was to examine the net gains from alternative

trading techniques that can be utilized by currency managers working for international banks and

hedge funds when trading the EUR/USD currency on an intraday basis. A buy and hold strategy,

sell and hold strategy, and a Bollinger Band strategy were applied to tick level sample data

gathered from 2009 to 2016 to determine net gains from each strategy. The results of an

ANOVA test indicate there is a statistically significant difference, however the Bollinger Band

strategy produced an overall net loss from trading. The findings suggest that using an alternative

trading strategy, Bollinger Bands, on an intraday basis does not increase net gains from trading

activity.

Key words: Currency trading, intraday, alternative trading, Bollinger Bands

AN EXAMINATION OF ALTERNATIVE TRADING TECHNIQUES USING INTRADAY

EUR/USD CURRENCY PRICES

by

Brock Vaughters

Doctoral Study Submitted in Partial Fulfillment

of the Requirements for the Degree of

Doctor of Business Administration

Liberty University, School of Business

August 2018

___________________________________________________

Dr. Rol Erickson, Dissertation Chair

___________________________________________________

Dr. Edward Moore, Dissertation Committee Member

___________________________________________________

Dr. Edward Moore, Program Director

___________________________________________________

Dr. David Calland, Dean, School of Business

Acknowledgments

I would like to began by thanking God for His amazing grace and love. It is for his

service that I have completed this journey. May He receive all praise and glory.

I would like to express my deepest gratitude to my wife, Angie, who has provided

encouragement and much needed proofreading these last several years. Also, I would like to

thank my children, Mia and Drake, who so understandingly sacrificed time with their Dad so that

he might fulfill the work that God has called him to complete.

I am indebted to the guidance, support, direction, and grace provided to me by my

committee chair, Dr. Rol Erickson. A doctoral candidate could not ask for a better chair to his or

her guide on this arduous journey. Additionally, the committee members, Dr. Edward Moore

and Dr. Gene Sullivan, have provide much need feedback and clarity along this path. I am truly

grateful for this amazing team of scholars and teachers.

Finally, I would like to express my appreciation for my colleagues in the Falls School of

Business at Anderson University. The support from Dean Terry Truitt throughout this entire

process has been nothing short of remarkable. Thank you to all of you for making this journey

possible.
List of Tables ................................................................................................................................. vi

List of Figures ............................................................................................................................... vii

Section 1: Foundation of the Study..................................................................................................1

Background of the Problem ................................................................................................ 1

Problem Statement .............................................................................................................. 3

Purpose Statement ............................................................................................................... 4

Nature of the Study ............................................................................................................. 4

Discussion of method.............................................................................................. 4

Qualitative method .................................................................................................. 5

Mixed method ......................................................................................................... 5

Quantitative method ................................................................................................ 5

Discussion of design ............................................................................................... 6

Experimental and quasi-experimental......................................................... 6

Descriptive .................................................................................................. 7

Correlation .................................................................................................. 7

Summary of the Nature of the Study .................................................................................. 7

Research Questions ............................................................................................................. 8

Hypotheses .......................................................................................................................... 8

Theoretical Framework ....................................................................................................... 9

Efficient market hypothesis .................................................................................... 9

Adaptive market hypothesis .................................................................................. 10

Technical analysis ................................................................................................. 10

Discussion of relationships between theories and variables ................................. 12

i
Summary of the Conceptual Framework .......................................................................... 13

Definition of Terms........................................................................................................... 13

Assumptions, Limitations, Delimitations ......................................................................... 14

Assumptions.......................................................................................................... 15

Limitations ............................................................................................................ 15

Delimitations ......................................................................................................... 16

Significance of the Study .................................................................................................. 16

Reduction of Gaps................................................................................................. 16

Implications for Biblical integration ..................................................................... 17

Relationship to field of study ................................................................................ 18

Summary of the Significance of the Study ....................................................................... 19

A Review of the Professional and Academic Literature ................................................... 19

Foreign currency exchange market ....................................................................... 20

Spot market ........................................................................................................... 22

Intraday currency prices ........................................................................................ 23

Financial market theory ........................................................................................ 25

Traditional market theories ................................................................................... 27

Efficient market hypothesis .................................................................................. 28

Support of the EMH .................................................................................. 30

Challenging the EMH ............................................................................... 32

Adaptive market hypothesis ...................................................................... 35

AMH and EMH combined ........................................................................ 36

Summary of market theories ..................................................................... 38

ii
Currency trading techniques ................................................................................. 39

Fundamental analysis factors .................................................................... 39

The carry trade .......................................................................................... 41

Technical analysis and trading strategies .................................................. 43

Quantitative analysis ................................................................................. 45

Challenges of technical analysis ............................................................... 46

In support of technical analysis................................................................. 47

Types of technical trading analysis ....................................................................... 50

Variables in the study............................................................................................ 51

Buy and hold strategy ............................................................................... 53

Sell and hold strategy ................................................................................ 53

Summary of the literature review ......................................................................... 54

Transition and Summary of Section 1 .............................................................................. 55

Section 2: The Project ....................................................................................................................56

Purpose Statement ............................................................................................................. 57

Role of the Researcher ...................................................................................................... 57

Participants ........................................................................................................................ 58

Research Method and Design ........................................................................................... 58

Discussion of method............................................................................................ 58

Discussion of design ............................................................................................. 59

Summary of Research Method and Design ...................................................................... 61

Population and Sampling .................................................................................................. 61

Data Collection ................................................................................................................. 63

iii
Instruments ............................................................................................................ 63

Data collection techniques .................................................................................... 64

Data organization techniques ................................................................................ 64

Summary of Data Collection ............................................................................................ 65

Data Analysis .................................................................................................................... 65

Variables used in the study ................................................................................... 65

Bollinger Band Parameters ....................................................................... 66

Bollinger Band Entry and Exit.................................................................. 67

Buy and hold: Parameters and entry/exit. ................................................. 69

Sell and hold: Parameters and entry/exit .................................................. 69

Transaction size, transaction costs, and interest ....................................... 69

Hypotheses 1 ......................................................................................................... 70

Hypotheses 2 ......................................................................................................... 71

Statistical analysis ................................................................................................. 71

Summary of data analysis ..................................................................................... 73

Reliability and Validity ..................................................................................................... 73

Reliability.............................................................................................................. 74

Validity ................................................................................................................. 75

Internal validity ......................................................................................... 75

External validity ........................................................................................ 76

Summary of reliability and validity ...................................................................... 76

Transition and Summary of Section 2 .............................................................................. 76

Section 3: Application to Professional Practice and Implications for Change ..............................78

iv
Overview of the Study ...................................................................................................... 78

Presentation of the Findings.............................................................................................. 79

Descriptive Statistics ......................................................................................................... 80

Normality .......................................................................................................................... 83

Hypothesis 1...................................................................................................................... 83

Hypothesis 2...................................................................................................................... 87

Relationship of the hypotheses to research questions ........................................... 90

Additional findings ............................................................................................... 91

Summary of Findings ........................................................................................................ 93

Applications to Professional Practice ............................................................................... 94

Currency Manager and Currency Trader .......................................................................... 95

Biblical Implications ......................................................................................................... 95

Recommendations for Action ........................................................................................... 97

Recommendations for Further Study ................................................................................ 98

Reflections ........................................................................................................................ 99

Summary and Study Conclusions ................................................................................... 100

References ....................................................................................................................................102

Appendix A: Listing of all Days ..................................................................................................138

Appendix B: Normality Test by Year Using Alpha of 0.05 ........................................................140

Appendix C: ANOVA by Year ....................................................................................................141

Appendix D: t-Test when ANOVA was Significant....................................................................143

v
List of Tables

Table 1. Descriptive Statistics for All Eight Years .......................................................................80

Table 2. Year-by-Year Descriptive Statistics for each Trading Strategy .....................................81

Table 3. ANOVA ..........................................................................................................................84

Table 4. t-Test Comparison of Bollinger Band Strategy and Buy and Hold Strategy ..................85

Table 5. t-Test Comparison of Bollinger Band Strategy and Sell and Hold Strategy ..................88

vi
List of Figures

Figure 1. Relationships between theories and variables. ..............................................................12

vii
Section 1: Foundation of the Study

In this paper, the researcher examined the net gain and losses of currency transactions

using technical trading tools that international banks and hedge funds can employ to positively

impact currency trading activities. The currency transactions that were studied are concentrated

on the EUR/USD, a major currency pair as determined by trading volume and will be used in

conjunction with the technical trading tool known as the Bollinger Band. Usage of alternative

trading strategies by financial firms has been researched extensively in the stock market and

futures market (De Bondt & Thaler, 1985; Duvinage, Mazza, & Petitjean, 2013; Hong & Stein,

1999; Kudryavtsev, Cohen, & Hon-Snir, 2013; Lakonishok, Shleifer, & Vishny, 1994; Lo,

2004). Much of the existing research in the currency markets uses end-of-day prices or weekly

prices to determine the effectiveness of alternative trading strategies to produce positive net

gains in the foreign exchange market. However, the use of Bollinger Bands as a means to

increase the net gain of currency trades made by trading teams working for global financial firms

have been scarcely researched. Additionally, no previous research has been found which focuses

on both Bollinger Bands and intraday currency trading.

Background of the Problem

Global commerce continues to expand as many countries implement the fundamental

ideas of open markets and free trade (Cetorelli & Goldberg, 2012; Meschi, Taymaz, & Vivarelli,

2016). The global expansion has been aided by advances in technology and communication

systems that allow companies to access information from around the world almost

instantaneously (Edoho, 2013). A key factor that supports global commerce is a well-

functioning currency market (Opie & Dark, 2015). Specifically, currency markets provide a

mechanism for converting the profits made in a local currency into home country monies

1
(Allayannis, Ihrig, & Weston, 2001). Currency markets, country-specific monetary policy, and

international banks all work together to support the growth of global commerce by servicing

global companies and their needs to conduct business in local currencies (Chong, Chang, & Tan,

2014; Georgiadis, & Mehl, 2016). It is this need for local currencies that may expose

international firms to exchange rate risk (Siow, 2013).

Transaction risk, commonly known as exchange rate risk, is a major factor in determining

the exposure that companies have to currency volatility (Dhargalkar, 2015). Transaction

exposure is common to all international firms that enter into contracts that obligate the making of

payments or the receiving payments in foreign currencies (Khindanova, 2015). The movement

of currency values makes it difficult for a company to accurately predict their expenses or

liabilities in other countries and the profit that foreign subsidies earn in the originating country’s

currency (Song, 2015). Consequently, the fluctuation of foreign exchange rates should be

addressed by multinational corporations to reduce risk in global transactions.

The challenge for many international firms is that currency markets are essential in the

conducting business transactions but are also a source of volatility and risk for all participants

(Wong, 2016). Global financial institutions provide a mechanism for multinational corporations

to hedge against exchange rate risk via currency futures contracts and spot exchange rates

(Chien, Lee, Tai, & Liao, 2013). Global financial institutions can take on the role of broker,

dealer, and speculator. As a broker or dealer, financial institutions sell currency derivative

products, for a fee, to their clients who need to hedge against transaction risk (Kamau, Inanga, &

Rwegasira, 2015). International financial institutions and hedge funds also speculate in currency

markets as a means to increase profits of their existing currency positions (Aktan, Chan, Žiković,

& Evrim-Mandaci, 2013; Shen & Hartarska, 2013). The role of currency traders, whether they

2
are working for multinational financial institutions or hedge funds, is to engage in currency

activities that increase the net gain per currency transaction for their respective firms (Fiedor &

Holda, 2016; Kamau et al., 2015).

Currency traders working for international banks and hedge funds use both fundamental

analysis and technical analysis when engaging in currency transactions (Eiamkanitchat,

Moontuy, & Ramingwong, 2017). Fundamental analysis, in the context of currency trading,

involves macro-level factors such as interest rate movements, default risk, budget surpluses and

deficits, and government manipulation (Abby & Doukas, 2015; Bitvai & Cohn, 2015; Kim &

Song, 2014). Technical analysis or chart analysis is based on past price movements, volume, and

a number of specific trading rules (Metghalchi, Kagochi, & Hayes, 2014). Several researchers

(Kuang, Schroeder, & Wang, 2014; de Zwart, Markwat, Swinkels, & Van Dijk, 2009) show that

combining technical trading rules and fundamental analysis can improve the net gain of currency

trading activities. Technical analysis tools such as the Bollinger Bands can be used by currency

traders and may support increased net gains realized from their foreign exchange trading activity

(Coakley, Marzano, & Nankervis, 2016; Chen, Chen, & Chuang, 2014; Lubnau & Todorova,

2015).

Problem Statement

International financial institutions and hedge funds utilize multiple methods to support

the trading activities of their currency traders (Thinyane & Millin, 2011). Two of the most

common methods are fundamental analysis and technical analysis (Chen et al., 2014;

Eiamkanitchat et al., 2017). Several researchers have indicated that financial firms that utilize

alternative trading tools can potentially increase the net gain from trading activities (Savin,

Weller, & Zvingelis, 2007; Zarrabi, Snaith, & Coakley, 2017). Other researchers argue that net

3
gains generated using alternative trading techniques are inconclusive activities (Coakley et al.,

2016; Hsu & Taylor, 2013; Kuang et al., 2014; Neely & Weller, 2003). These contradictory

findings leave currency managers uncertain of the effectiveness of alternative trading strategies

to improve net gains from currency trading activities (Duvinage et al., 2013; El Ouadghiri &

Uctum, 2016; Galariotis, 2014).

The general business problem was that international financial institutions were not

utilizing alternative trading tools which could potentially result in net trading gains for the firm

due to the inconclusive evidence from researchers. The specific business problem was some

currency trading managers overseeing EUR/USD spot currency traders lack adequate data to

determine if alternative trading tools could increase net gains.

Purpose Statement

The purpose of this quantitative study was to examine the net gains from alternative

trading techniques that can be utilized by currency managers working for international banks and

hedge funds when trading the EUR/USD currency on an intraday basis. The results may provide

additional insights for currency managers with regards to the use of alternative trading strategies

on an intraday basis within the currency markets. The findings of the research may have a direct

application to the business problem of increasing the net gain from foreign currency transactions

undertaking by currency management teams and currency traders within global financial

institutions and hedge funds. Also, this examination will add to the current literature on

technical analysis and help fill in the gap with regards to intraday time frames.

Nature of the Study

Discussion of method. There are three core research methods available for the research

to choose from: the qualitative method, the quantitative method, and the mixed method. When

4
conducting a study, the researcher needs to identify and select the research methodology that will

align with and support the research questions (Yin, 2006). Therefore, determining the proper

research method at the beginning of this study helps to ensure that all the information supports

the research questions.

Qualitative method. Qualitative research is a method that has a “reliance on human

perceptions and understanding” (Stake, 2010, p. 11). Scholars using qualitative research often

attempt to explain a particular social behavior or way of thinking (Yin, 2011). Multiple sources

of data can be collected in the qualitative research approach to explore the specific context that

may help explain a particular observed behavior (Creswell, 2014). Researchers using a

qualitative study approach seek to contribute insights from existing concepts or attempts to

develop new concepts (Yin, 2011). Qualitative research was not an appropriate research method

for the problem identified because analyzing personal experiences would not be appropriate to

address the research questions.

Mixed method. The mixed method approach incorporates both qualitative and

quantitative research methods. Researchers employing a mixed method approach attempt to

provide a more thorough and complete understanding of the research problem by those

employing this method (Stake, 2010). By combining both methods, the researcher hopes to gain

greater clarity than either the qualitative or the quantitative approach can provide on their own

(Creswell, 2014). The mixed method approach was not an appropriate research method for the

problem identified in this study. The qualitative portion of the mixed method would not be

appropriate to address the research questions.

Quantitative method. Researchers using quantitative studies seek to examine various

hypotheses by testing the relationship between numerical variables (Creswell, 2014).

5
Quantitative researchers tend to gather objective, historical, numerical sample data and then

analyze these sample sets using statistical tools. Researchers then gain insights based on the

results of the statistical analysis (Stake, 2010; Yin, 2011). The quantitative research method

supports the nature of the data collected for this study. More importantly, since the purpose of

this research is to determine if there is any correlation between net gain within foreign exchange

transactions and various trading techniques, the quantitative research method will be used

(Creswell, 2014).

Discussion of design. There are multiple research designs used in quantitative research:

experimental, quasi-experimental, descriptive, and correlation. Selecting the appropriate

research design is important to ensure the research question is addressed. Choosing a proper

research design will depend upon the problem to be addressed, the research questions, and the

nature of the data available (Parylo, 2012).

Experimental and quasi-experimental. Researchers using experimental and quasi-

experimental design strive to conclude if a specific treatment (independent variable[s]) can

influence the result (dependent variable) of a specific test. Experimental research can be

accomplished when a researcher provides a specific treatment to one group and withholds the

treatment from another group to determine how both groups performed during the experiment

(Creswell, 2014). Experimental design is the only research approach that can support a direct

cause and effect relationship given that researchers can manipulate and control certain variables

in the testing (Cash, Stankovic, & Storga, 2016). Experiments can further be broken down into

quasi-experiments, which use non-randomized sample sets to make inferences about the

population, and true experiments, which use randomized sample sets to make inferences about

the population (Keppel, 1991). The key factor in experimental design is the use of a control

6
variable. Due to the inability to control and manipulate independent variables, neither the

experimental design nor the quasi-experimental design was selected for this study.

Descriptive. Descriptive research design within quantitative methods seeks to describe

and explain what is occurring. Surveys and other observational data gathering tools are often

used in descriptive statistics, but no attempt is made to control or influence any variables

(Spector, Merrill, Elen, & Bishop, 2013). This research design was not selected given that the

variables and data used in this study have already been defined and described.

Correlation. Correlation research design typically uses sample data gathered from a

population to determine if there is any relationship between the variables being tested (Cash et

al., 2016). Additionally, within correlation studies, there is no attempt by the researcher to

control any variables used in the study (Meyers, Gamst, & Guarino, 2016). Sample data for this

study were gathered from foreign currency markets over an eight-year period, and no attempt to

control any of the variables in the sample set were made. The dependent variable, the net gain

from trades, has been documented by several other researchers (Menkhoff & Taylor, 2007;

Metghalchi et al., 2014; Bitvai & Cohn, 2015). Three different trading techniques were the

independent variables. These techniques were applied to the data to determine if a specific

trading technique had a higher positive correlation with net gains and to determine if there was a

statistical difference between the different group means. Given the nature of the data and

problem statement, conducting a correlation study was justified as the purpose of the study was

to determine if there was a relationship between net gain and two different trading strategies.

Summary of the Nature of the Study

In summary, the focus of this research is to examine if currency trading managers at

international banks and hedge funds experience a greater net gain from currency transactions

7
when employing a traditional trading approach (long and short strategies) as compared to an

alternative trading strategy when trading the EUR/USD spot market on an intraday period. A

quantitative study was selected for this study to examine the relationship between net gain from

currency trading and alternative trading strategies using Bollinger Bands. Net gains from trading

will be the dependent variable. The traditional trading strategy of buy and hold, a strategy of

shorting the EUR/USD, and the alternative trading strategy will be the independent variables.

Net gains as a dependent variable are documented by other researchers (Menkhoff & Taylor,

2007; Metghalchi et al., 2014; Bitvai & Cohn, 2015). These variables relate directly to the

specific problem to be addressed in this study, which is some managers of currency traders are

unclear if using alternative trading strategies on an intraday basis can increase the net gains from

currency trading.

Research Questions

Research Question 1: Do currency trading managers at international banks and hedge

funds experience a difference in net gain from currency transactions when employing a

traditional trading approach as compared to an alternative trading strategy when trading the

EUR/USD spot market on an intraday period?

Hypotheses

Null and Alternative Hypotheses for Research Question 1:

H10: There is no statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the traditional buy and hold strategy.

8
H1a: There is a statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the traditional buy and hold strategy.

H20: There is no statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the sell and hold strategy

H2a: There is a statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the sell and hold strategy.

Theoretical Framework

According to Creswell (2014), the theoretical framework for research studies is used to

provide a foundation for which the research can be built. The two theoretical frameworks found

in the financial literature that will serve as the foundation for this quantitative correlation study

are (a) the efficient market hypothesis and (b) the adaptive market hypothesis. The modern

financial market theory is based on the seminal work of Eugene Fama who developed the

efficient market hypothesis (EMH) in 1970 (Fama, 1970). More recently, Andrew Lo (2004) has

proposed the adaptive market hypothesis (AMH) which includes elements of behavioral finance

as a means to explain prices, volatility, and profits in financial markets.

Efficient market hypothesis. The premise of the efficient market hypothesis (EMH) is

that the current price of a financial asset reflects all publicly available information and that

engaging in any speculative trading activity in an efficient financial market should produce a net

gain of zero (Westerlund & Narayan, 2013). The EMH lists three distinct forms, or states of the

market. The weak-form of the EMH assumes that the current price of a financial asset already

9
includes all historical information about that asset and market participants cannot achieve

consistent excess returns using technical analysis (Kofarbai & Zubairu, 2016). Financial assets

prices in a semi-strong efficient market will reflect all known historical and all currently

available information (Bitvai & Cohn, 2015). Finally, the strong-form of the EMH states that the

value of an asset reflects all information to include historical price information, current public

information, and non-public information (Fama, 1970). The strong form indicates that it is

impossible to earn excess profits from the market even using insider information (Degutis &

Novickyte, 2014). The EMH concludes that no individual can consistently outperform market

returns when risk, trading costs, and luck are appropriately factored (Duvinage et al., 2013). For

this research, I will study the weak-form of the EMH by examining a specific correlation

between historical price information and net gains in the foreign currency market. According to

the weak-form of the EMH, the net gain from using historical price information should be zero.

The results of this study might show evidence that is contra to the weak-form EMH.

Adaptive market hypothesis. A growing number of research articles introduced certain

behavioral elements as a way of explaining the financial market activity (Lo, 2004). Various

behaviors manifest themselves via different individual investment approaches when trading

financial instruments (Duvinage et al., 2013). The main behavioral approaches have focused on

momentum investing (Hong & Stein, 1999); contrarian behaviors (Lakonishok et al., 1994);

behavior biases (Kudryavtsev et al., 2013); news-driven behavior (De Bondt & Thaler, 1985);

and over-reaction or under-reaction (Hong & Stein, 1999). Many of these behaviors are

identified by traders using technical analysis tools.

Technical analysis. Technical analysis uses historical price data and other historical

market data to predict future price movement (Coakley et al., 2016). Various technical trading

10
tools are commonly used by currency traders as a means to increase net gains for their respective

companies (Neely, Weller, & Ulrich, 2009). It can be reasoned that even the slightest advantage

of just a few pennies, or even a few seconds, can provide currency traders with a significant

advantage in the market over time (Narayan, Mishra, Narayan, & Thuraisamy, 2015). That is

why these predictive tools are so critically important to the financial sector and currency traders

specifically (Antinolfi & Kawamura, 2008).

Furthermore, extensive academic research has been devoted to determining the

effectiveness of the various predictive tools (Bitvai & Cohn, 2015). Much attention has been

given to the study of these mathematical models for two primary reasons. The first is due to the

enormous sums of money that can be made in the currency markets if there is a tool that can

increase net gains. The second reason is that if traders can earn excessive risk-adjusted profits in

the financial markets, this would provide significant evidence against the efficient market

hypothesis (Neely et al., 2009).

The combination of the efficient market hypothesis, specifically the weak-form, and the

adaptive market hypothesis, specifically technical analysis, provides the theoretical support for

this study. Currency managers working for global financial institutions and international hedge

funds should use every method at their disposal to increase the net gains from foreign currency

transactions as a means to maximize shareholder value (Morck, 2014). These efforts should

include the use of alternative trading techniques if those techniques prove to increase the net gain

from trading activities.

11
Dependent Net Gains
Variable

Efficient Market Adaptive Market

Theory Hypothesis Hypothesis

Buy and Hold Technical

Independent Strategy
Sell and Hold
(Decrease in
Analysis (looking
(Increase in for patterns in
Variables value)
value)
the market)

Figure 1. Relationships between theories and variables.

Discussion of relationships between theories and variables. The premise of the

efficient market hypothesis (EMH) is that the current price of a financial asset reflects all

publicly available information and that engaging in any speculative trading activity in an

efficient financial market should produce a net gain of zero (Westerlund & Narayan, 2013).

Therefore, the dependent variable selected for this study was net gain from trading activity.

Additionally, the weak-form of the EMH assumes the current price of a financial asset already

includes all historical information about that asset and market participants cannot achieve

consistent excess returns using technical analysis. Based upon this understanding of the weak-

form of the EMH one independent variable was the technical analysis tool known as Bollinger

Bands. Two more independent variables that directly relate to the EMH were the buy and hold

strategy and the sell and hold strategy. The buy and hold type strategy represented those market

12
participants who believed values will increase, and the sell and hold strategy represented those

market participants that take a contrarian view of the market believing values will decrease.

Summary of the Conceptual Framework

The combination of the efficient market hypothesis, specifically the weak-form, and the

adaptive market hypothesis, specifically technical analysis, provides the theoretical support for

this study. Currency managers working for global financial institutions and international hedge

funds should use every method at their disposal to increase the net gains from foreign currency

transactions as a means to maximize shareholder value (Morck, 2014). These efforts should

include the use of alternative trading techniques if those techniques prove to increase the net gain

from trading activities.

Definition of Terms

Alternative Trading Strategies: This refers to trading tools and strategies that attempt to

“time the market” with regards to entry and exit trades. This strategy relies on technical trading

tools to assist with decision-making (Coakley et al., 2016).

Bollinger Bands: This technical trading tool consists of three bands. The middle band is

based upon the moving average of the currency pair, and the lower band and the upper band are

two standard deviations from the moving average (Abby & Doukas, 2012).

Bubbles: Financial anomalies that represent extended deviations from rational

fundamental valuations of assets (Brock, Hommes, & Wagener, 2009; Phillips & Yu, 2011).

Buy and Hold Strategy: An investment strategy where an asset is purchased with the

belief the value of the asset will increase over time creating a net gain (Cohen & Cabiri, 2015).

13
Forex or FX: These are abbreviations for the foreign currency exchange market which

operates as an over-the-counter market without a centralized clearing exchange (Neely et al.,

2009).

Net Gain: The total aggregate realized profits, inclusive of transaction costs, from trading

using either alternative trading strategies or traditional buy and hold strategies (Bitvai & Cohn,

2015).

Noise Trader: A financial market trader who makes buy and sell decisions based upon

idiosyncratic information as opposed to fundamental or economic information (Aabo, Pantzalis,

& Park, 2017; Li, 2016).

Over-the-counter market: A highly decentralized financial market consisting of dealers

that simultaneously buy and sell assets to investors (Duffie, Gârleanu, & Pedersen, 2007; Lester,

Rocheteau, & Weill, 2015).

Short and Hold Strategy: An investment strategy where an asset is sold to the market at a

high price with the belief that the value of the asset will decrease over time and can be bought

back at a lower price creating a net gain (Mohamad, 2016).

Tick-level data: Financial data recorded every time there is a change in price regardless

of the timeframe. In currency trading, tick-level data are gathered any time there is a price

change of 0.000001 or more (Drachal, 2016; Kuo, Chen, & You, 2017).

Assumptions, Limitations, Delimitations

“Bias, the lack of objectivity, is by definition a predisposition to error…” (Stake, 2010, p.

164). The following is a discussion of the assumptions, limitations, and delimitations used in

this research project. This information is key to providing readers with an additional perspective

on the findings of the study (Yin, 2011).

14
Assumptions. Assumptions made in this research study include the accuracy of

historical intraday data examined. The Forex market is a globally decentralized over-the-counter

market (Ackerman, 2016; El Ouadghiri & Uctum, 2016). This means that there is no single

source for all currency data transactions. Data for this study were obtained from a third party

data provider, TickData, LLC. TickData, LLC overcomes the issues of decentralization and

over-the-counter execution by aggregating currency quotes from several contributing institutions.

Each contributor provides primary, secondary, and tertiary data. This process provides data that

is isolated, redundant and geographically diverse. Through a proprietary data system, TickData,

LLC compares all three levels of currency exchange prices and synthesizes all information into a

single data item free from errors and omissions (TickData, 2017). Additionally, the parameters

used to test the Bollinger Band tool are assumed to be consistent throughout the period examined

in the study. Finally, the research questions for this study sought to determine if there is a

correlation between the use of Bollinger Bands and an increase in net gains. Given that

Bollinger Bands are based on historical price data, the study assumes the weak-form of the

efficient market hypothesis is not constantly valid and the adaptive market hypothesis is

applicable.

Limitations. The findings of the study are limited to the time period tested. Sample data

for this study will be gathered from January 2009 through December 2016 for the currency pair

of EUR/USD. Any statistical difference that might be identified in this study between the buy

and hold strategy, sell and hold, and the Bollinger Bands tool, may not carry over to future time

periods or other currency pairs. The confidence level chosen for this study is 90% and the

confidence interval is 10% based upon the sample size. Given these levels, the results from the

sample may not be generalizable to the entire population. Finally, there is a limited number of

15
research studies that have tested intraday currency data. Although there is a great deal of

academic research that utilizes daily prices and various technical tools together, there are none

that are known to this researcher which have used intraday price changes to test technical

indicators.

Delimitations. There are hundreds of currencies traded within the Forex markets (Fong,

2013). For this study, the EUR/USD currency pair, one of the most common currency pairs, was

selected. According to the Bank of International Settlement, the USD and the EUR account for

the largest portion of exchange turnover volume in the spot market (April, 2016). The

EUR/USD was chosen because of the significant volume that these two currencies represent in

the overall Forex market.

Both the euro and United States dollar are all currencies that are free from government

intervention or manipulation (Satterlee, 2014). Stated differently, the currencies chosen for this

study are free-floating currencies allowed to fluctuate with the supply and demand of currency

market participants. Free floating currencies will provide results that have a minimal influence

of individual monetary policy objectives of specific nation states. The time frame selected for

this study was 2008 through 2012. This period includes the global financial crisis, the recession

that followed, and the beginning of economic recovery. The timeframe was selected to provide a

robust understanding of the correlation between trading techniques in a variety of market

conditions.

Significance of the Study

Reduction of Gaps. Currently, there seems to be a gap in the literature relating to

intraday trading and technical analysis (Ozturk, Toroslu, & Fidan, 2016). Much of the current

literature focuses on the end of day price changes of currency pairs and does not focus on

16
intraday price changes (Hayes et al., 2016; Narayan et al., 2015; Poti et al., 2014). Additionally,

most studies that have tested Bollinger Bands and other reversion-to-the-mean technical trading

tools have focused on individual stocks, ETFs, or indices without much focus on specific foreign

currency trading (Brock, Lakonishok, & LeBaron, 1992; Lento, Gradojevic, & Wright, 2007;

Duvinage et al., 2013; Metghalchi et al., 2014). In this study, the researcher attempted to add to

the scholarly body of knowledge by focusing on intraday price changes and the application of

Bollinger Bands to currency trading.

Implications for Biblical integration. As a believer in the teachings of the Bible, it is

important to integrate the understanding of stewardship into all areas of life. The biblical view

of stewardship is traditionally defined as using and managing the resources God provides for the

glory of God and the betterment of his creation (Botha, 2014). Stewardship can further be

understood as entrusting someone with resources and acting on behalf of another’s interest

(McCuddy & Pirie, 2007). Currency traders working for hedge funds and international banks

have a responsibility to make decisions that are in the best interest of the business owners and

clients of the firm (Morck, 2014). By utilizing the best tools available, currency traders can

provide the greatest level of value maximization to clients and owners (Osiyevskyy &

Biloshapka, 2017). This project is focused on increasing the net gains from currency trading

activities and therefore engaging in the practice of stewardship by ensuring the greatest net gains

for clients and owners.

Finally, the role of the global financial institutions must fit into God’s design. “There are

different kinds of service, but the same Lord. There are different kinds of workings, but the

same God works all of them for all men” (1 Corinthians 12:5-7, NLT). Global financial traders

and business professionals can fulfill God’s purpose by providing resources for God’s creation to

17
flourish with the understanding that these goods and services are beneficial to God’s people.

Positive net gains are vital to any for-profit organization, but within God’s design, making

money provides a means to serve God’s creation rather than being the ultimate pursuit of the

organization (Van Duzer, 2010). All people are responsible for the resources they control, and

wasted resources quickly yield no fruit (McCuddy & Pirie, 2007).

Ensuring good stewardship within the foreign exchange markets for international banks

and hedge funds includes the exploration of alternative investment methods. These

organizations need to make sure they are employing the most effective tools that will maximize

the return for shareholders, who can then use these funds to further the work of God on Earth.

Relationship to field of study. Increased global commerce has given rise to the

importance of foreign currency exchange markets (Harvey, 2013). Multinational firms engage in

business activities that are outside of their home country and must adapt products, services, and

pricing to each local market to be successful (Devereux, Dong, & Tomlin, 2017). The local

currency is the easiest way for local consumers to make purchases from international firms

(Satterlee, 2014). These global companies then rely on the services of international banks to

exchange the foreign currency for the currency of the home country. Due to the growth of

globalization, the foreign exchange market has increased exponentially. The foreign exchange

market (Forex or FX) is the largest financial market in the world with an estimated 5.3 trillion

dollars of turnover every single day (Menkhoff, 2015). All trading takes place between the

world’s largest banks, who act as market makers, and all other market participants (NASDAQ,

2016). This large, liquid market also attracts hedge fund traders and investors looking to

increase their net gains (Fong, 2013). Currency trading is an essential function of all

18
international businesses, a key service for international banks, and a vital source of profitability

for international hedge funds.

Summary of the Significance of the Study

Currently, there seems to be a gap in the literature relating to intraday trading and

technical analysis (Ozturk et al., 2016). Much of the current literature focuses on the end of day

price changes of currency pairs and does not focus on intraday price changes (Hayes et al., 2016;

Narayan et al., 2015; Poti et al., 2014). In this study, the researcher attempted to add to the

scholarly body of knowledge by focusing on intraday price changes and the application of

Bollinger Bands to currency trading. Increasing the net gains from currency trading activities

engages the practice of stewardship by ensuring the greatest net gains for clients and owners.

Finally, due to the growth of globalization, the foreign exchange market has increased

exponentially. Currency trading is an essential function of all international businesses, a key

service for international banks, and a vital source of profitability for international hedge funds.

A Review of the Professional and Academic Literature

The focus of this study is to examine the relationship between net gains and alternative

trading techniques that can be utilized by currency managers working for international banks and

hedge funds. There are five elements to this literature review. First, this literature review

contains a review of the existing structure of currency markets. Second, this literature review

contains an analysis of the related research as it pertains to market theories which will serve as

the theoretical basis for this study. This portion includes a comparison and contrast of the

different scholarly findings regarding the ability of financial market participants to increase the

net gains from trading activity. Third, this literature review contains a concise summary of the

relevant aspects about currency trading techniques and which trading strategies have been found

19
to increase net gain from trading activity. Many of these strategies and trading tools have been

tested in various types of quantitative studies. Fourth, this literature review presents support for

the use of the research variables of net gains and alternative trading strategies used in this study.

Finally, this literature review is based on scholarly peer-reviewed journals and sound academic

articles.

The literature review contains an extensive review of scholarly work about currency

trading, market efficiency, currency markets, and various market analysis techniques. The

scholarly peer-reviewed literature includes academic journal articles, with the vast majority

being published within the last five years. Books, textbooks, whitepapers, and conference

materials were also included as part of this literature review. This literature review contains

approximately 299 different sources, and over 75% of the citations used in this literature review

have been published within that last 3-5 years. The volume of literature on this topic provides an

indication the level of interest in the financial markets of both academics and practitioners.

Foreign currency exchange market. An exchange rate is defined as the price of one

currency stated in terms of another currency (Chadwick, Fazilet, & Tekatli, 2015). The foreign

exchange market is the most heavily traded financial market in the world with roughly $5.3

trillion exchanged daily (Evans, Pappas, & Xhafa, 2013; Zarrabi et al., 2017). According to

Coakley et al. (2016), the global currency market is highly liquid and daily transaction volume is

“several times greater than the combined transaction volume of largest stock exchanges” (p.

273). The foreign currency market also referred to as FX or Forex, is a decentralized, over-the-

counter, worldwide market that operates 24 hours a day, five days a week (Ackerman, 2016; El

Ouadghiri & Uctum, 2016). The over-the-counter Forex market is an unregulated market, unlike

a regulated exchange like the New York Stock Exchange. Instead, oversight for the Forex

20
market occurs via independent market makers and various worldwide electronic networks

(Baranga, 2016; Powers, 2016). Financial centers around the world serve as a marketplace for

transactions between various buyers and sellers, and most transactions occur between private

parties (Ackerman, 2016; Powers, 2016). However, not all currency transactions occur equally

distributed around the world. The Bank of International Settlements (2001, 2004, 2007, 2010)

reports that nearly 50% of all currency turnover or volume, occurs in New York and London

giving these financial centers considerable influence in the currency markets (Harvey, 2013).

There are two distinct levels of trading within the Forex market. The first level is

considered retail trading (Rime & Schrimpf, 2013; Sager & Taylor, 2006). Retail trading takes

place between a financial institution and an individual. Individuals, in this case, did not refer to a

person, but rather a non-bank which could be a government entity, corporation, institutional

investors, hedge funds, or high net worth clients (Rime & Schrimpf, 2013; Sager & Taylor,

2006). The next tier of FX transactions is the wholesale level which occurs between financial

institutions (Bleaney & Li, 2016; Powers, 2016). Banks that have a large number of orders to

buy or sell a currency can leverage the interbank dealer network to satisfy their client demands

(Bleaney & Li, 2016; Powers, 2016). Central banks take on the role of market makers and must

be ready to buy and sell currencies from member banks to ensure the Forex market functions

smoothly (Krapl & Giaccotto, 2015; Langfield & Soramäki, 2016). Currency traders working

for international banks or hedge funds will trade currencies in the near-term on the spot Forex

market, as opposed to the FX futures market.

The foreign exchange market plays an important role in global financial markets.

Currency traders working for hedge funds and global financial institutions find considerable

volume and liquidity in the FX market. High transaction volume and significant liquidity both

21
help currency traders enter and exit positions on an intraday basis. Currency traders can engage

in currency transactions in the futures market or the spot market (Bleaney & Li, 2016; Kraple &

Giaccotto, 2015; Powers, 2016). The focus of this study will be on the spot exchange market.

Next is a discussion of the spot currency market examining some of the spot markets unique

characteristics.

Spot market. Trading in the spot currency market is done by buying and selling

currency pairs (Talebi, Winsor, & Gavrilova, 2014; Pintar et al., 2016). Currency traders

speculate, or place trades, on the changing price of currencies between two countries.

Throughout the entire trading period that the FX market is open one currency will fall (rise) as

another currency rises (falls) in relation to each other. The FX spot market, also known as the

cash market, is the currency market for immediate delivery with all trades settling within two

days of the transaction (Powers, 2016; Rzepczynski, 2008). The spot market does not include

trades of financial instruments but instead is an exchange of one currency for another at the

current exchange rate or spot rate (Gerber, 2016; Kamau et al., 2015). In contrast, the futures

market includes the same current pairs, but the exchange will take place on some date in the

future (Pintar et al., 2016). The spot market is used by corporations, government entities, and

financial institutions to speculate currency trades and to hedge currency risks. The speculation

and hedging activities by these significant global institutions are what provide the transaction

volume and liquidity in the FX market (Akerman, 2016; Powers, 2016; Talebi et al., 2014).

Spot currency trading includes two currencies that are simultaneously traded (Ackerman,

2016; Pintar et al., 2016). Simultaneously trading two currencies at the same time is known as

pairs trading. Powers (2016) provides an example of a spot currency transaction where Bank X

agrees to sell 100 million euros to Bank Y for U.S. dollars at a spot rate of 1.10. In this example,

22
Bank X will receive 110 U.S. dollars by supplying 100 million euros. The process of spot

currency trading is roughly identical to the mechanics found in all other markets except for the

fact that FX traders are always short one currency and simultaneously long in another

(Abildgren, 2014; Powers, 2016). Finally, the spot FX market is self-regulated and no single

agency has authority over it. Much of the regulation of the FX market occurs as a byproduct of

oversight and laws that govern banks in each country (Ackerman, 2016; Talebi et al., 2014).

This lack of unified oversight and the sheer size of the FX market makes the Forex unique as

compared to all other financial markets (Ackerman, 2016; Baranga, 2016; El Ouadghiri &

Uctum, 2016).

The spot currency market for currency contracts that will be settled in the near term,

typically within two days. The volume in the spot market is provided by global financial

institutions, governments, international corporations, and speculators. These individuals trade

the spot market for different reasons but each is concerned with the very short-term and intraday

currency prices. Understanding intraday currency price movement is an important aspect to this

research study.

Intraday currency prices. Currency traders and currency managers make their trading

decisions over multiple time horizons to include annually, monthly, daily, hourly, and down to

the minute level (Abbey & Doukas, 2015; Hsu, Taylor, & Wang, 2016). Many financial experts

support the premise that fundamental economic data does not provide any predictive power in

the short-term but have found evidence of exchange rate predictability over a several year period

of time (Andersen, Bollerslev, Diebold, & Vega , 2007; Boudoukh, Richardson, & Whitelaw,

2008). The most significant issue with attempting to predict short-term (second-by-second)

changes in exchange rates with fundamental economic data is the economic data does not change

23
that quickly (Dal Bianco, Camacho, & Perez Quiros, 2012, Mcmillan & Speight, 2012). New

economic data are typically released on a monthly, quarterly, or annual basis as compared to

currency prices that fluctuate on a second by second basis (Boudoukh et al., 2008; McMillan &

Speight, 2012).

As a means to better predict the price movement of currency pairs, previous researchers

have used various statistical methods and tick-level-data to identify intraday periodic patterns

based upon the macroeconomic news releases (Andersen & Bollerslev, 1997; Barunik, Krehlik,

& Vacha, 2016).

The study of intraday price movements is relatively new (Rosch, Subrhmanyam, & Van

Dijk, 2017). Tick-level data are the recording of price changes as they occur in the market

(Curato & Lillo, 2014). In contrast, intraday currency prices on a minute time frame records the

currency exchange rate every minute. Tick-level data are gathered anytime there is a price

change regardless of the time frequency (Bréhier, 2013). This information was not readily

obtained nor was there computing power available to process vast amounts of financial data

before the year 2000 (Hau, Massa, & Perress, 2010; Mende & Menkhoff, 2006; Ranaldo, 2009;

Rosch et al., 2017). Previous researchers have focused on the day of the week effect in Forex

markets (Ke, Chiang, & Liao, 2007; Ito & Hashimoto, 2006; Serbinenko & Rachev, 2010), but

few researchers examined the foreign exchange market on an intraday basis. Popovic and

Durovic’s (2014) study examined intraday pricing anomalies between the USD and EUR

currencies and discovered that there is a profitable arbitrage opportunity on Friday afternoons.

Elaut, Frömmel, and Lampaert (2016) researched intraday patterns in the Russian Ruble and

revealed that trading momentum was likely driven by risk aversion to holding Rubles overnight.

Kablan and Ng (2011) studied intraday FX prices to develop an adaptive neuro-fuzzy interface

24
for learning to predict currency movements at the tick level. Several studies find order flow

occurring in intraday currency prices to detect patterns in trader behavior Ben Omrane &

Hussain, 2016; Breedon & Ranaldo, 2013; Fuertes, Kalotychou, & Todorovic, 2015; Rime,

Sarno, & Sojli, 2010). Many of the conclusions reached by the researchers previously cited

support the premise that currency traders making trading decisions on an intraday basis have the

potential to increase net gains from trading activity. There are several articles that study the

Forex market on an intraday basis. However, no previous research has specifically examined the

intraday day price movement of the EUR/USD and Bollinger Bands. This gap in the literature is

intriguing given the fact that many practitioners make trading decisions throughout the trading

day and not just based on the end of day prices.

Currency traders and managers should understand the foreign exchange market to

compete with other traders and increase their net gains from trading (Akerman, 2016; Powers,

2016; Talebi et al., 2014). Understanding the market conditions, market participants, operating

procedures, timeframes for trading, and the overall function of the market is essential to

successful trading (Hau et al., 2010; Mende & Menkhoff, 2006; Ranaldo, 2009; Rosch et al.,

2017). Additionally, for currency traders to increase their net gain from trading, they must be

able to accurately predict the movement or direction of currency prices in the near term (Elaut et

al., 2016; Kablan & Ng, 2011; Popovic & Durovic, 2014). Given the importance of accurately

predicting price movements, the next section presents several financial market theories

developed over the past half-century that have attempted to explain the movement of asset

prices.

Financial market theory. Financial markets present a mechanism to facilitate the

movement of capital from savers to borrowers (Brigham & Houston, 2015). These markets can

25
provide traders with an opportunity to make significant profit and net gains for those traders who

can correctly predict the direction of financial markets (Harvey, 2013; Poti & Siddique, 2013).

Two main financial theories have been developed to explain the movement of financial markets.

These two financial theories found in the financial literature that will serve as the foundation for

this quantitative correlation study are (a) the efficient market hypothesis and (b) the adaptive

market hypothesis. The efficient market hypothesis and the adaptive market hypothesis are the

two prevalent market theories that attempt to explain how financial markets move (Lo, 2004;

Rosch et al., 2017; Talebi et al., 2014). The modern financial market theory is based on the

seminal work of Eugene Fama who developed the efficient market hypothesis (EMH) in 1970

(Fama, 1970). More recently, Andrew Lo (2004) has proposed the adaptive market hypothesis

(AMH) which includes elements of behavioral finance as a means to explain prices, volatility,

and net gains in financial markets. These EMH and the AMH were selected for this study

because these market theories directly relate to the purpose of this study and the research

questions.

Other market related theories exist but do not directly relate to the research question or

purpose of this study. Portfolio theory, for example, examines the appropriate mix of various

investment assets to produce the highest return for a given level of risk (Fu & Blazenko, 2015;

Fu & Blazenko, 2017; Hou, Xue & Zhang, 2014). The Arbitrage Pricing Theory (APT) attempts

to explain the financial returns, or net gains, of assets as a function of the risk factors and also

considers how financial markets price these factors (Huberman, 1982; Rebeschini, & Leal,

2016). The APT is used exclusively in equity markets and was therefore not appropriate for this

study, as the researcher examined the currency markets. The put-call parity is a financial theory

that deals directly with various European style options, their prices, and the value of the

26
respective underlying asset (Cerreia-Vioglio, Maccheroni, & Marinacci, 2015; Plott, &

Pogorelskiy, 2017; Stoll, 1973). The put-call parity deals specifically with option prices and is

therefore not directly related to the research questions nor the purpose of this study. After

reviewing these various theories, the researcher has determined that the EMH and the AMH are

the most relevant theories to the research questions and the purpose of this study.

Traditional market theories. The EMH indicates that market participants cannot

achieve net gains above the overall market returns in the long run because a developed financial

market efficiently disseminates information such that there is no sustainable financial advantage

allowing individuals to earn excess returns (Fakhry, 2016; Fama, 1970). Andrew Lo (2004), first

proposed the AMH. Lo theorized that markets are not in a constant state of efficiency. The

AMH is constructed on three findings. First, financial market efficiency is not a steady state, and

therefore alternative or technical trading rules can be employed to achieve significant net gains.

Second, competition and currency traders desiring net gains will drive market participants to

learn and change over time. These “adaptive” processes will gradually erode profitable trading

opportunities over time. Finally, more complex technical trading strategies should persist longer

than simple trading strategies that can be replicated by market participants (Coakley et al., 2016).

The EMH presents an argument that currency traders cannot earn excess market returns.

The AMH indicates that traders can earn excess returns but can do so at a diminishing rate as

other market participants erode profitable trading strategies over time. In this literature review,

the researcher provided a basis of support for further study of the weak form of the EMH and the

AMH using previous academic research. The researcher intends to add to the body of scholarly

literature regarding the AMH and the weak form of the EMH in the intraday currency markets.

By examining the relationship between alternative trading strategies and net gains, the findings

27
of this study might lend support for the AMH. If there is no correlation between historical prices

using alternative trading strategies and net gains in the currency markets, then this might support

the EMH. Because the findings of this study might support either the AMH or the EMH, both of

these theories play an important role in the conceptual framework of this study. The next

sections discuss the EMH and the AMH in greater detail.

Efficient market hypothesis. The EMH is used as part of the therortical framework for

this study. The EMH implies that the net gains from attempting to forecast future asset prices

would result in zero profits and is not consistently forecastable (Chen & Diaz, 2013; Fleming &

Remolona, 1999; Malkiel, 2005; Mobarek & Fiorante, 2014; Timmermann & Granger, 2004). In

this study the researched examined if there is a correlation between historical currency price

movements and net gains from trading activity. The findings presented in this study may be

contra to the EMH. The findings may indicate that historical prices can be used to increase net

gains from trading in the currency market.

The premise of the efficient market hypothesis (EMH) is that the current price of a

financial asset reflects all publicly available information and that engaging in any speculative

trading activity in an efficient financial market should produce a net gain of zero over time (Hsu

et al., 2016; Westerlund & Narayan, 2013; Kofarbai & Zubairu, 2016). The EMH indicates that

prices of financial assets will immediately reflect the fundamental value of the asset.

Additionally, the EMH assumes a highly competitive market where participants are profit

maximizing, risk-minimizing, and 100 percent rational individuals (Fakhry, 2016). There was

early empirical evidence for the EMH. Jensen (1978) stated “there is no other proposition in

economics which has more solid empirical evidence supporting it than the efficient market

hypothesis” (p. 167). The EMH implies that market participants cannot realize profits in excess

28
of overall market return and future prices are not forecastable from historical prices (Chen &

Diaz, 2013; Fleming & Remolona, 1999; Malkiel, 2005; Mobarek & Fiorante, 2014;

Timmermann & Granger, 2004).

The EMH lists three distinct forms, or states, which include (a) the weak-form, (b) the

semi-strong form, and (c) the strong-form (Fakhry, 2016; Jensen, 1978; Malkiel & Fama, 1970).

The weak-form of the EMH assumes the current price of a financial asset already includes all

historical information about that asset and market participants cannot achieve consistent excess

returns using technical analysis (Kofarbai & Zubairu, 2016). Financial asset prices in a semi-

strong efficient market should reflect all known historical and all currently available information

(Bitvai & Cohn, 2015, Kristoufek & Vosvrda, 2016). Therefore, based on the EMH, market

participants cannot realize profits from using historical or current information (Popovic &

Dorovic, 2014; Rosch et al., 2017). The semi-strong form of the EMH indicates that investors

and traders cannot earn a return that exceeds the market consistently over time (Chen & Diaz,

2013; Manahov & Hudson, 2014; Mobarek & Fiorante, 2014).

Finally, the strong-form of the EMH states that the value of an asset reflects all

information to include historical price information, current public information, and non-public

information (Fama, 1970). The strong-form indicates that it is impossible to earn excess profits

from the market even using insider information (Degutis & Novickyte, 2014; Manahov, Hudson,

& Gebka, 2014). The EMH concludes that no individual can consistently outperform market

returns when risk, trading costs, and luck are appropriately factored (Duvinage et al., 2013;

Manahov & Hudson, 2014). According to Fakhry (2016), the EMH indicates that the financial

markets consist of market participants that are perfectly rational, risk-averse and profit-

maximizing at all times. Given the rational, risk-averse, profit-maximizing participants, the goal

29
of traders should not be to attempt to outperform the market but instead attempt to maximize the

level of return given the level of risk (Bitvai & Cohn, 2015; Malkiel, 2005). Eugene Fama was

the first to put forth a concise theory regarding the movement of the market. Since 1970, when

Fama proposed the EMH, many researchers provided substantial evidence for and substantial

evidence against the EMH.

Support of the EMH. Around the 1900s, the French mathematician named Louis

Bachelier (1900) observed the stock market prices follow a Brownian motion. This Brownian

motion is better known today as randomness or the random walk. Fama (1963, 1965, 1970, and

1995), measured the statistical properties of market prices and concluded that market prices

follow a random walk pattern. From this evidence, Fama proposed the three forms of the EMH

that is (a) the weak-form, (b) the semi-strong form, and (c) the strong form. The foundation of

the EMH is that a financial market is largley comprised of rational, profit-maximizing, risk-

averse, and well-capitalized investors (Manahov & Hudson, 2014; Tsang, 2017). Additionally,

the current market price of assets is firmly grounded in the fundamental value of those assets and

prices only experience change when new information about the fundamental value is introduced

to traders (Fiebig & Musgrove, 2015; Popovic & Dorovic, 2014). Several researchers, analyzing

the price movement of various financial assets such as U.S, equities, international equities,

ETF’s, mutual funds, commodities, and futures, have found evidence in support of the EMH

(Jawadi, Jawadi, & Cheffou, 2015; Kumar, 2013; Majumder, 2013; Mobarek & Fiorante, 2014;

Papadamou & Markopoulos, 2014; Thompson, 1978).

Thompson (1978) and Brauer (1988) found evidence in support of the strong-form of the

EMH researching closed-end mutual funds indicating that prices tend to move in a random

fashion and traders could not consistently outperform the market return over the long-term.

30
Other researchers have conducted studies in the options market (Galai, 1978), the futures market

(Oliven & Rietz, 2004; Westerlund & Narayan, 2013; Westerlund, Norkute, & Narayan, 2015),

the U.S. equity markets (Chen & Diaz, 2013; Ito, Noda, & Wada, 2016; Narayan, 2006), the

commodities market (Kristoufek & Vosvrda, 2016; Papadamou & Markopoulos, 2014) and

various international equity markets (Jawadi et al., 2015; Kumar, 2013; Majumder, 2013;

Mobarek & Fiorante, 2014) and found similar random price movements indicating the existence

of an efficient market. These researchers have used various statistical methods to vary the

randomness of the price movements in each of these markets. The findings of these researchers

lend support to the conclusion that markets are in one of the three forms of the EMH as described

by Fama (1970).

The focus of this research is to examine if currency markets are efficient, specifically

weak-form efficient as described by the EMH. The weak-form EMH presents the conceptual

framework for testing the research questions by stating that historical pricing data are not

effective in helping currency traders increase net gains from trading activity. Researchers

concluded that markets will operate efficiently over the long run. In this research, the researcher

examined intraday prices over a longer period of several years. Additionally, researchers

conclude that markets operate efficiently if the pricing data appears to follow a random pattern.

These random pricing patterns imply that currency traders cannot consistently outperform the

market return. In this study, the researcher examined if currency prices can help currency traders

predict near-term price movements.

In this section reviewing the literature in support of the EMH, there appears to be

considerable support early on for the theory, from 1970 until early 2000. More recently, scholars

are finding evidence against the EMH and tend to challenge the EMH instead of support EMH.

31
The following section will examine the scholarly research that challenges many of the

assumptions of the EMH.

Challenging the EMH. The foundation of the efficient market hypothesis (EMH) is that

the current price of a financial asset reflects all publicly available information and asset prices

follow a random pattern (Fama, 1970). If prices are random and asset prices reflect all available

information, then engaging in any speculative trading activity in an efficient financial market

should produce a net gain of zero over time (Hsu et al., 2016; Westerlund & Narayan, 2013;

Kofarbai & Zubairu, 2016). However, Grossman and Stiglitz (1976, 1980), question the premise

of informationally efficient markets from a theoretical and practical viewpoint. Grossman and

Stiglitz found that market participants typically pay for information and therefore prices cannot

perfectly reflect all information (1976, 1980). Kay (2013) stated that if market prices reflect all

available information about the value of an asset, traders and investors would have no incentive

to obtain that information in the first place because they could observe it in the price of the

financial asset. Various researchers (Condie & Gangulie, 2011; Kovalenkov & Vives, 2014;

Mele & Sangiorgi, 2009; Vivies, 2014) have attempted to answer the Grossman-Stiglitz paradox

by creating a variety of theoretical models dealing with noise traders and finite population

assumptions. Many researchers have agreed that they could not overcome the fact that market

participants do spend resources obtaining information and rely on this information in making

trading decisions (Kovalenkov & Vives, 2014; Vivies, 2014). This practice tends to violate the

instantaneous information symmetry of the EMH (Condie & Gangulie, 2011; Grossmand &

Stiglitz, 1980).

Another important assumption of the EMH is rational behavior on the part of market

participants (Fama, 1970). Various behavior patterns, other than entirely rational behavior, have

32
been observed in financial markets. These various behaviors present themselves via different

individual investment approaches when trading financial instruments (Duvinage et al., 2013;

Nofer & Hinz, 2015). The primary behavioral approaches have focused on (a) momentum

investing (Foltice & Langer, 2015; Hong & Satchell, 2015; Hong & Stein, 1999), (b) contrarian

behaviors (Lakonishok et al., 1994; Sobaci, Sensoy, & Erturk, 2014; Stefanescu & Dumitriu,

2016), (c) behavior biases (Chang, Huang, Chang, & Lin, 2015; Jacobs & Hillert, 2016;

Kudryavtsev et al., 2013), (d) news-driven behavior (De Bondt & Thaler, 1985; Lachanski &

Pav, 2017), and (e) over-reaction or under-reaction (Hong & Stein, 1999; Kleinnijenhuis,

Schultz, Oegema, & Atteveldt, 2013). These researchers have observed that significant

movement in asset prices can be attributed to behavior that can be considered irrational (Nofer &

Hinz, 2015). Therefore, the findings presented in these studies are contrary to the assertion of

the EMH that individuals act rationally. In summary, many scholars find support of irrational

market participant behavior which is inconsistent with the assumptions of the EMH.

Other researchers have questioned whether financial markets are mainly efficient with

periods of inefficiency or are financial markets mainly inefficient with periods of efficiency

(Neely, Weller, & Dittmar, 2009; Pukthuanthong, Levich, & Thomas, 2007)? Based upon these

continually changing and competing forces “convergence to equilibrium is neither guaranteed

nor likely to occur at any point in time” (Tsang, 2017, p. 468). In this way, markets vary over

time, and the efficiency of markets vary over time, never existing in a constant continuous state

of efficiency (Alvarez-Ramirez, Rodriguez, & Espinosa-Paredes, 2012; Menkhoff & Taylor,

2007; Menkhoff, Sarno, Schmeling, & Schrimpf, 2016; Neely et al., 2009; Pukthuanthong et al.,

2007). These findings lend support to the idea that financial markets are always moving back

and forth between states of efficiency and inefficiency.

33
Researchers have found evidence that the Forex market deviates substantially from

information efficient markets and rational market behavior as outlined in the EMH (Levich &

Poti, 2015; Neely et al., 2009; Poti & Siddique, 2013). Beginning in the 1970s, around the

inception of the EMH, empirical evidence has shown that various technical trading rule can

result in statistically significant net gains in currency markets (Dooley & Shafer, 1975; Popovic

& Durovic, 2014; Poti, Levich, Pattitoni, & Cucurachi, 2014; Poti & Siddique, 2013). There is a

large number of research studies that presented evidence of statistically significant profits in the

currency markets by using simple technical trading rules such as moving average crossovers and

support and resistance (Chang & Osler, 1999; LeBaron, 1999; Schulmeister, 2006).

Furthermore, in developing nations, which do not have laws ensuring systematic, efficient

information flow and a large number of well-capitalized investors, there are signs of

predictability of price movements (Menkhoff et al., 2016; Nwachukwu & Shitta, 2015). These

factors provide traders the opportunity to earn excess profits (Nwachukwu & Shitta, 2015) above

the typical market return. Interesting to note that researchers have found evidence of

diminishing profitability of some technical trading rules over time (Menkhoff & Taylor, 2007;

Menkhoff et al., 2016; Neely et al., 2009; Pukthuanthong et al., 2007) lending support for the

idea that markets change, and traders are continually adapting to changing markets, economic

conditions, change sentiment, and then reacting to the actions of other traders.

The existing research on the subject of EMH is vast and diverse. Interestingly, even the

developer of EMH, Eugene Fama (1995) and supporters (Malkiel. 2003, 2005; Jensen, 1978)

acknowledge that all financial markets can have short periods of time where traders are not

entirely rational and predictable patterns can emerge. Examples of such irrationality occur in

bubbles, market crashes, and recessions. Andrew Lo (2004, 2005), in an effort to accommodate

34
the research findings that support EMH and the empirical evidence questioning the validity of

the EMH, proposed a new theory known as the adaptive market hypothesis (AMH).

Adaptive market hypothesis. Lo (2004) suggested that the AMH is a means to reconcile

the EMH with the observed behavioral biases in financial markets. He suggests that markets are

complex, adaptive, and influenced continuously by individual bias, competition, selfishness, and

other environmental factors (Lo, 2004). Lo was not the first to recognize how markets adapt

over time. The term adaptive efficiency was first used by Kent Daniel and Sheridan Titman

(1999). These authors observed investor over-confidence or investor irrationality in the U.S.

equity markets and determined that profits could be made from the momentum generated by

over-confidence. However, the financial market displayed an adaptive efficiency whereby the

profit opportunities diminished as more and more traders realized the opportunity (Daniel &

Titman, 1999). Adaptive efficiency referred to the observation of pricing anomalies that

disappear as more and more investors take advantage of the discrepancy (Daniel & Titman 1999;

Menkhoff & Taylor, 2007). These researchers also found evidence of abnormal and persistent

net gains from specific investor portfolios which should not be possible under the EMH (Cohen

& Hon-Snir, 2013; Daniel & Titman, 1999; Stefanescu & Dumitriu, 2016). The findings of these

researchers indicate that there are opportunities for currency traders to increase net gains from

trading activity. This lends support to the conceptual framework of this study as the researchers

provide evidence of the possibility of net gains from trading activity in the financial markets.

Others scholars (Lo, 2012; Kay 2013; Vives 2014) question the ability of well-capitalized

arbitrage traders to counter the irrational behavior of market participants. These authors find

support for the belief that well-capitalized investors typically join into market exuberance and

market crashes as demonstrated in market booms and market busts. The findings of these

35
researchers are important to the conceptual framework for this study as they provide evidence of

the possibility of net gains from trading activity in the financial markets.

The AMH is based on four essential concepts (Lo, 2002; Tsang, 2017). First, market

efficiency is not a steady state, and therefore technical trading rules can be employed to achieve

significant profits (Lo, 2004; Urquhart & Hudson, 2013). Second, competition and the desire for

profits will drive market participants to learn and change over time. These “adaptive” processes

will gradually erode profitable trading opportunities over time, but give an opportunity for new

profitable trading strategies. More recent studies have shown that these simple trading rules are

not as profitable as they once were (Neely et al., 2009). Advanced technical trading rules,

however, have proved to produce statistically significant profits in currency markets in the recent

past (Coakley et al., 2016). These findings lend support to the Adaptive Market Hypothesis

(AMH) proposed by Lo (2004). Third, the relationship between risk and reward is not universal

from one investor to the next and that each will define optimal net gains differently (Dhankar &

Shankar, 2016; Poti et al., 2014). Finally, more complex technical trading strategies should be

able to return positive net gains for extended periods of time as compared to simple trading

strategies that can be quickly learned by market participants (Coakley et al., 2016; Levich &

Poti, 2015).

AMH and EMH combined. As previously mentioned, many researchers have found

evidence in support of the EMH while other researchers have found support for the AMH. Still,

some researchers believe that there is room for a co-existence of the AMH and the EMH and that

these theories are not mutually exclusive. The AMH alters the EMH and Lo further suggests that

the forces of learning, competition, and evolutionary selection pressures drive prices to their

efficient level (Manahov & Hudson, 2014; Shalini, 2012). These same forces then create new

36
profitable trading opportunities until they are discovered and traders begin to adapt.

Additionally, Lo (2004, 2005) theorizes the rational, profit-maximizing, well capitalized, and

risk-averse investor is not always present or dominant in financial markets. The AMH offers a

mechanism to reconcile the empirical findings of researchers regarding the predictability of asset

returns with the concepts of the EMH (Levich & Poti, 2015; Shalini, 2012). That is, the AMH

helps merely to explain the anomalies that other researchers have observed regarding the

predictability of asset returns, but over time market efficiency will eliminate these sources of

profit (Charles, Darne, & Kim, 2012; Pintar et al., 2016; Pukthuanthong et al., 2007), but

changing conditions will create new opportunities, and the cycle starts over again.

Hiremath and Kumair (2014) agreed with the basic foundations of the AMH but advise

that further study is necessary given the relative newness of the theory. These authors conclude

that empirical studies must be performed as a means of building the necessary academic support

for such a hypothesis. It is essential to keep in mind that there is no formal statistical test that

can directly test the AMH (Hiremath & Kumair, 2014; Noda, 2016). These are not criticisms of

the AMH, but rather serve as cautionary observations which lead to a conclusion of “wait and

see” before declaring the AMH as the singular market hypothesis which can explain the price

patterns of financial assets (Coakley et al., 2016; Levich & Poti, 2015).

The researchers that recently tested the AMH concluded that the AMH provides an

accurate description of market behaviors (Ghazani & Araghi, 2014; Hull & McGroarty, 2014;

Manahov & Hudson, 2014; Urquhart & Hudson, 2013; Verheyden, Van den Bossche, & De

Moor, 2015; Zhou & Lee, 2013). A long-term study of 100 years of U.S. stock market data finds

that efficiency has dramatically increased after the 1980s (markets have adapted) and specific

market conditions (recession, expansion, market crashes which indicate irrational behavior) are

37
significant factors in predicting returns (Kim, Shamsuddin, & Lim, 2011). Alvarez-Ramirez et

al. (2012) found the period of highest efficiency for the U.S. stock market was between 1973 and

2003, but changed over time. Charles et al. (2012) observed periods of predictability in the

Forex market for developing countries and periods of market efficiency, or randomness, appear

to be present in markets of developed nations. The financial market in Japan has shown signs of

moving from periods of efficiency to periods of inefficiency and back (Noda, 2016). These

findings support the AMH hypothesis by showing how markets change over time and how

competition among traders diminish the profitability of various strategies and provide

opportunities for new trading strategies.

Summary of market theories. The problem addressed in this quantitative study was to

examine historical currency data to assess whether or not using alternative trading strategies on

an intraday basis can increase the net gains from currency trading. Given the previous findings

(Darne & Kim, 2012; Ghazani & Araghi, 2014; Hull & McGroarty, 2014; Manahov & Hudson,

2014; Urquhart & Hudson, 2013; Verheyden et al., 2015; Zhou & Lee, 2013), it appears that

financial markets experience periods of predictability. If markets do experience some level of

predictability, these findings provide a strong theoretical foundation for the research in the area

of increasing net gain currency trading. The researcher examined the weak-form of the EMH to

determine if there is any correlation between historical price information and net gains in the

foreign currency market. The strength of the correlation between past prices and net gains would

help support other researchers who have found evidence against the EMH. Additionally, the use

of historical prices to increase net gain beyond the market returns for currencies would provide

support against the weak-form of the EMH. According to the weak-form of the EMH, the net

38
gain from using historical price information should be zero. The results of this study might show

evidence that is contra to the weak-form EMH.

Currency trading techniques. The foreign exchange market attracts a vast number of

traders and investors who speculate in the exchange rates to maximize their net gain (Addam,

Chen, Hoang, Rokne, & Alhajj, 2016). Each currency trader makes decisions based on several

qualitative and quantitative factors. The diversity of factors considered and the weight given to

each factor differ from one trader to the next (Abuhamad, Mohd, & Salim, 2013; Beilis, Dash, &

Wise, 2014). These differences further support the AMH, indicating that traders are adapting the

level of weighting for each factor based on previous learnings (Lo, 2004). Many of the factors

that influence a trading decision will fall into two broad categories (a) fundamental analysis

factors and (b) technical analysis factors (Chen et al., 2014; Coakley et al., 2016; Ozturk et al.,

2016). These categories are not mutually exclusive, and most traders use a combination of

several factors from both fundamental and technical analysis (Chang, Jong, & Wang, 2017;

Kuang et al., 2014; Levich & Poti, 2015; Manahov et al., 2014). The use of fundamental factors

or technical analysis factors is meant to help traders enhance their predictive power of the Forex

market (Georgiadis & Mehl, 2016; Hayes et al., 2016; Jacobs & Hillert, 2016). The following is

a review of some of the more common factors that currency traders consider when trading the

Forex market.

Fundamental analysis factors. Fundamental analysis is considered the conventional

approach to assessing the potential for future gains and loss of various investments

(Eiamkanitchat et al., 2017). Fundamental analysis is based on historical financial and economic

data as well as projections of future demand. Fundamental factors tend to be macroeconomic

factors that analyzes the overall state of a country’s economy (Dobrynskaya, 2015;

39
Eiamkanitchat et al., 2017). The fundamental factors that tend to impact foreign currencies are

projected GDP growth, inflation, interest rates, and the growth of money balance (Abuhamad et

al., 2013; Sarno & Schmeling, 2014). These fundamental factors typically impact entire

countries and regions, not just individual sectors or firms. Currency managers adhering to a

strictly fundamental analysis approach to foreign currency trading would focus solely on

macroeconomic factors that affect each country and then take positions that would align with

these variables (Copeland & Lu, 2016; de Zwart et al., 2009; Dobrynskaya, 2015). Currency

managers using fundamental analysis typically take a long-term view, from three to five years.

The three to five-year time horizon is necessary because macroeconomic variables typically do

not change to a great extent and often change slowly over multiple years.

Previous researchers examined the correlation between macroeconomic factors and short-

term currency trading. Bekiros (2014) studied the connection between exchange rates and

predictability of future rates using fundamental variables of money supply and interest rates.

Bekiros determined that the linkage between fundamentals and predictability is modest and not

uniform across time periods. Earlier studies (Belaire-Franch, & Opong, 2005; Flood & Rose,

1995; Meese & Rogoff, 1982; Rogoff & Starvrakeva, 2008) agreed with the findings of Bekiros

and added that currency rates seem to follow a random walk pattern when correlated with

macroeconomic factors, although not consistently. Bacchetta and Van Wincoop (2013) found

that the macroeconomic fundamentals change so slowly over time that these variables do not

provide value for predicting currency movements in the near-term. The authors did find that it is

the expectation of a change in these variables that causes movement in the underlying currency

(Bacchetta & Van Wincoop, 2013). However, the announcement of fundamental factors does

seem to impact currency trading in the short-term.

40
New information and news announcements about changes in interest rates, inflation, and

net foreign asset positions are associated with substantial and quick currency movements (Dal

Bianco et al., 2012; Dobrynskaya, 2015). Andersen, Bollerslev, Diebold, and Vega (2003),

Andersen et al. (2007), and Faust, Rogers, Wang, and Wright (2007) found that fundamentals do

not appear to help forecast currency price movements in the short run, but news about

fundamentals does create quick and significant short-term movements in currency exchange

rates. Ultimately, the issue with using fundamental economic data, outside of news related

events, is the economic data are not available at the same frequency rate as price data (Copeland

& Lu, 2016; de Zwart et al., 2009; El Ouadghiri & Uctum, 2016). The findings of these

researchers imply that short-term currency trading based on economic fundamentals may not

produce a positive net gain consistently over time. There does appear to be one trading strategy,

known as the carry trade, that is based upon the fundamental factor of interest rates that have

been shown to produce profitable results over the short-term (Coudert & Mignon, 2013; Fung,

Tse, & Zhao, 2013; Hutchison & Sushko, 2013).

The carry trade. One relevant theory in international finance is the interest rate parity

theory (Cenedese, Sarno, & Tsiakas, 2014; Kim, 2016). Based on the interest rate parity theory,

traders created the carry trade strategy to generate positive net gains from currency transactions

(Burside et al., 2007; Coudert & Mignon, 2013). The interest rate parity theory indicates that

there should be an equilibrium relationship between the movement of exchange rate pairs and

their respective interest rates (Baillie & Chang, 2011; Fung et al., 2013; Mollick & Assefa,

2013). Kim (2016) stated “the currency with a higher nominal interest rate is expected to

depreciate against the other currency by roughly the same amount as the interest rate differential

in the absence of transaction costs” (p. 1077). According to Kim (2016), interest rate parity

41
works in theory but in practice appears to work in the exact opposite manner. Empirical

evidence has shown the interest rate parity works in the opposite direction where higher yielding

currencies appreciate over time instead of depreciating (Anzuini & Fornari, 2012; Burside et al.,

2007; Coudert & Mignon, 2013). Currency traders can take speculative trades by purchasing

high yield currencies while simultaneously borrowing in low yield currencies when interest rate

differences are not offset by corresponding interest rate movements (Bakshi & Panayotov, 2013;

Hutchison & Sushko, 2013). This strategy is known as the currency carry trade (Das,

Kadapakkam, & Tse, 2013; Lustig, Roussanov, & Verdelhan, 2014). Researchers have reported

statistically significant positive net gains from carry trades in the Japanese Yen (Colavecchio,

2008), the Japanese Yen and the Swiss Franc (Mollick & Assefa, 2013), and many of the G10

currencies (Jurek, 2014). Finally, Burnside, Eichenbaum, and Rebelo (2008) find that carry

trades can increase the Sharpe Ratio of a hedge fund by more than 50%. As with any speculative

market activity, the potential for loses is present (Das et al., 2013; Pan, Tang, & Xu, 2016). The

potential for loss in a carry trade strategy arises when high levels of exchange rate volatility

happen unexpectedly (Menkhoff, Sarno, Schmeling, & Schrimpf, 2012). Unexpected volatility

can occur from events stemming from surprise election results, unexpected monetary policy

changes, or in developing nations government coups and takeovers (Coudert & Mignon, 2013;

Fung et al., 2013; Hutchison & Sushko, 2013. These events can create significant losses from

exposed positions in carry trades.

In summary, fundamental analysis in the Forex market involves the study of historical

financial and economic data as a means to project future economic factors (Copeland & Lu,

2016; de Zwart et al., 2009; Dobrynskaya, 2015). The fundamental factors that are most often

studied include those macro-level indicators that affect an entire economy such as projected GDP

42
growth, monetary supply, inflation, interest rates, surplus and deficits, and the growth of money

balance (Abuhamad et al., 2013; Sarno & Schmeling, 2014). Researchers studying these factors

as a means to forecast future currency prices have yet to successfully find high correlations

between macro-level economic data and short-term currency price movements. Except for the

carry trade, the findings of these researchers implies that short-term currency trading based on

economic fundamentals may not produce a positive net gain consistently over time (Burnside et

al., 2008; Colavecchio, 2008; Jurek, 2014; Mollick & Assefa, 2013).

Given the results that have been presented, many researchers have turned their attention

to quantitative and technical analysis. The use of technical analysis can be a divisive topic and

often confounds economist, and traditional financial theorist as technical analysis directly

contradicts the weak-form EMH (Narayan et al., 2015; Neely et al., 2009). Technical analysis is

a key part of this study and as such a thorough discussion is necessary. The following is a

review of the academic literature regarding technical analysis and the use of technical analysis in

financial markets.

Technical analysis and trading strategies. Technical analysis is the study of financial

market activity through the application of price charts and empirical rules set mainly by active

market participants to predict future price movements (Milionis & Papanagiotou, 2013; Neely et

al., 1997). Technical analysis focuses primarily on historical price data and other historical

market data to predict asset prices in the future (Chang et al., 2017; Coakley et al., 2016). Due to

the importance of accurately forecasting short-term currency movements, extensive academic

research has been devoted to determining the effectiveness of the various predictive tools (Bitvai

& Cohn, 2015; Chang et al., 2014). Additionally, technical analysis has become an essential

aspect for financial market practitioners. Most major brokerage firms, international financial

43
firms, hedge funds, and even individual market professionals use technical analysis as a means to

understand the movements of financial assets and assist in forecasting potential direction of

future price movements (Brock et al., 1992; Zarrabi et al., 2017). Much attention has been given

to the study of technical trading for two primary reasons. The first is due to the importance of

financial markets and currency markets to practitioners, academics, and global economies

(Chang et al., 2014; Lo, 2002). Secondly, if traders can earn excessive risk-adjusted profits in

the financial markets, this would provide significant evidence against the efficient market

hypothesis (Lo, 2002; Narayan et al., 2015; Neely et al., 2009). Evidence for or against a market

theory, whether the EMH or the AMH, moves the understanding of all market participants

forward and furthers practicioners understanding of market functions.

The first scholarly study of technical analysis was titled “Can Stock Market Forecasters

Forecast?” written in 1933 by Alfred Crowles who studied whether 45 stock professionals were

able to predict future price movement. However, the use of technical analysis for trading is

believed to date back to the rice trade in Japan as early as the 1700 (Lin, Yang, & Song, 2011;

Neely, Rapach, Tun, & Zhou, 2014; Northcott, 2009). Many believe that these rice traders were

first to develop a technical strategy known as Japanese Candlesticks (Chen, Bao, & Zhou, 2016;

Romeo, Joseph, & Elizabeth, 2015).

In the United States, Charles Dow, a market practitioner and founder of the Dow Jones

Financial News Services in the late 1800s, used the closing prices of stocks as a means of

predicting future price movement (Brock et al., 1992; Lin et al., 2011; Nazario, Lima-eSilva,

Amorim-Sobreiro, & Kimura, 2017). The writings of Charles Dow form the foundational

concepts of technical analysis today (Oliveira, Nobre, & Zárate, 2013; Vanstone & Finnie, 2009;

Zhu & Zhou, 2009). Technical analysis covers a broad category of tools and rules that traders

44
use to help interpret and predict future price movement. The intensive and exclusive use of

technical analysis has created a whole new trading style that is known as quantitative analysis.

Quantitative analysis. Quantitative analysis, commonly viewed as a subset of technical

analysis, uses mathematical and statistical tools to help investors discover hidden patterns in

financial data (Davis, 2017; Gerlein, McGinnity, Belatreche, & Coleman , 2016; Kirilenko, Kyle,

Samadi, & Tuzun, 2017). This branch of technical analysis focuses on the use of machine

learning, neural networks, and support vector machine techniques to discover patterns in large

amounts of financial data (Arnoldi, 2016; Kampouridis & Otero, 2017; Manahov et al., 2014;

Singh & Srivastava, 2017). Quants, as they are referred to in the industry, use these quantitative

techniques to create new technical tools to identify the most suitable moments throughout the

trading day to open and close positions (Huang, Hsu, Chen, Chang, & Li, 2015; Manahov,

Hudson, & Hoque, 2015; Wang, Smith, & Hyndman, 2008). These newly created quantitative

tools typically rely on small statistical advantages in the market, but the net gains are increased

by executing potentially thousands of trades throughout a trading session (Gerlein et al., 2016;

Wang et al., 2008). This study focused on the use of a common technical analysis tool.

However, the researcher did not develop new quantitative indicators using advanced machine

learning. Therefore, quantitative analysis, as defined here, was not a focus of this study. The

researcher, instead, focused on traditional and pre-existing technical analysis tools as they might

be applied to the Forex market.

The foreign currency markets often employ a great deal of technical analysis (Chang et

al., 2017; Zarrabi et al., 2017). It is estimated that up to 40% of FX traders around the globe use

some technical analysis to assist in their trading (Kolkova, 2017; Manahov et al., 2014). The use

of technical analysis often confounds economist, and traditional financial theorist as technical

45
analysis directly contradicts the weak-form EMH. Menkhoff and Taylor (2007) propose the

reason for the widespread use of technical analysis is that fundamental economic factors such as

interest rates and GDP are effective at explaining the long-term exchange rate movements;

however, these variables are far less useful at explaining the exchange rate movement within the

short-term and intraday basis. Blume, Easley, and O’Hare (1994) stated “because technical

analysis helps traders interpret current information, watching the sequence of market statistics

allows traders to correctly update their beliefs” (p. 177).

Challenges of technical analysis. There is considerable academic research that does not

support the use of technical analysis as an effective tool to produce positive net gains in trading.

Park and Irwin (2010) and Marshall, Cahan, and Cahan (2008a, 2008b) examined the use of

technical analysis in the commodity futures market and found no statistical support for the

profitability of technical analysis. Other researchers (Anderson & Faff, 2005; Bajgrowicz &

Scaillet, 2012; Sullivan, Timmermann, & White, 1999) found that technical analysis was not

effective in generating profits in the S&P 500 index or the Dow Jones Industrial Average over

various time periods. Goldbaum (2003) found that technical trading rules can be profitable, but

the fluctuation in popularity of various technical trading rules leads to losses in the long term.

Several researchers discovered that technical trading rules supported the generation of

profits (Coe & Laosethakul, 2010; Hudson, Dempsey, & Keasey, 1996; Zhu et al., 2015).

However, when those profits were compared to the profits that would have been generated using

the traditional buy and hold strategy, the researchers determined that technical analysis could not

generate excess profits above the buy and hold strategy (Coe & Laosethakul, 2010; Hudson et

al., 1996; Zhu, Jiang, Li, & Zhou, 2015). Additional researchers have found that various

technical trading tools to support profitable trading in excess of the buy and hold strategy. Yet,

46
when these researchers accounted for the transaction costs associated with the technical trading

signals, the technical analysis did not outperform the buy and hold strategy (Ellis & Parbery,

2005; Frömmel & Lampaert, 2016; Zakamulin, 2014).

The findings of these researchers provide two critical aspects to consider when examining

the results of any study using technical analysis (a) excess profits above the buy and hold

strategy, (b) transaction costs. Technical analysis might produce positive net gains, but those net

gains must exceed the net gains realized from the buy and hold strategy (Coe & Laosethakul,

2010; Hudson et al., 1996; Zhu et al., 2015). Additionally, the study must incorporate

transaction costs as technical analysis will generate more trading costs than a buy and hold

strategy (Ellis & Parbery, 2005; Frömmel & Lampaert, 2016; Zakamulin, 2014). Therefore,

profits from technical analysis must be higher than the buy and hold strategy after accounting for

transactions costs.

In support of technical analysis. Scholars often discount the use of technical analysis

because it contradicts an entirely rational approach to the financial markets as outlined by the

EMH (Zoicas-Ienciu, 2016). However, over the past several decades there is significant

academic research that supports the use of technical analysis in various financial markets. In

emerging markets, technical analysis has been shown to produce excess profits beyond the

traditional buy and hold strategy (Fifield, Power, & Knipe, 2008; Liu, Ji, & Jin, 2016; Ni, Liao,

& Huang, 2015; Shynkevich, 2017; Sobreiro et al., 2016). Technical trading rules support

traders in predicting price movements in emerging markets better than those in developed

nations (Zhu et al., 2015), which might lend support for the EMH by indicating that developing

nations do not have efficient markets and developed nations do have efficient markets (Liu et al.,

2016; Zhu et al., 2015). Researchers also found that technical traders in the market help to

47
reduce volatility by quickly identifying pricing errors and capitalizing on these errors

(Detollenaere & Mazza, 2014; Chiarella & Ladley, 2016).

Extensive research has focused on a primary technical trading strategy known as the

moving average technical trading strategy (Bonenkamp, Homburg, & Kempf, 2011; Han, Hu, &

Yan, 2016; Liu et al., 2017). Researchers applied the moving average cross-over strategies to

various equity markets and found these strategies were able to generate profits that exceeded the

buy and hold strategy and accounted for the additional transactions cost associated with technical

trading (Ahmad et al., 2017; Glabadanidis, 2014, 2015, 2017). Researchers have found similar

results in the commodities market (Clare, Seaton, Smith, & Thomas , 2014; Han et al., 2016; Liu

et al., 2017; Szakmary, Shen, & Sharma, 2010) using moving average and momentum studies.

Bonenkamp et al. (2011) deviated from traditional research studies by combining the use of

fundamental analysis (free cash flow) in equity markets and technical analysis to form a trading

strategy that exceeds the transaction costs and outperforms the buy and hold strategy. These

studies lend support for implementing technical analysis in trading as a means to increase net

gains.

Earlier findings from researchers presented evidence that technical analysis applied to

foreign currency markets could outperform a buy and hold strategy (Cornell & Dietrich, 1978;

Sweeney, 1986). These early results were further supported by researchers analyzing various

currency pairs and timeframes (Chang & Osler, 1999; Gencay, Ballocchi, & Dacorogna, Olsen,

& Pictet, 2002; LeBaron, 1999; Levich & Thomas, 1993; Neely et al., 1997; Schulmeister, 2006,

2009). Recent technical analysis studies in the foreign currency market have shown that profits

have diminished over time (Neely et al., 2007; Olson, 2004; Pukthuanthong et al., 2007). The

current research into technical analysis and the Forex market tends to combine various types of

48
technical analysis-like order flow (Ghyselse, Hill, & Motegi, 2013; Gradojevic & Lento, 2015)

and momentum trading (He & Li, 2015; Manahov, Hudson, & Gebka, 2014; Narayan et al.,

2015; Orlov, 2016). Gradojevic and Gencay (2013) find that “fuzzy” technical trading rules can

outperform simple moving averages in the EUR/USD currency, particularly on high volatility

trading days, and produce significant profits. The findings of these researchers support the use

of technical analysis in the Forex market.

In summary, the importance of accurately forecasting short-term currency movements

has been recognized by both practitioners and academics. Evidence of this importance can be

found in the considerable resources devoted to determining the effectiveness of the various

predictive tools. Technical analysis is one of these predictive tools and usually involves

examining financial market activity through the application of price charts and empirical rules.

Although some researchers have found evidence against the profitability of technical analysis in

the Forex market, a majority of scholars conclude that using technical analysis in the foreign

exchange market can produce statistically significant net gains for currency traders. The findings

of these scholars suggest that technical trading is a way for currency portfolio managers to

exploit short-term misprices in the market (Poti et al., 2014). Additionally, technical analysis

became an essential aspect for financial market practitioners (Neely et al., 2007; Zhu et al.,

2015). Most international financial firms and hedge funds engaged in the Forex market use

technical analysis as a means to understand the movement of currency prices.

For this research, the researcher studied the use of technical analysis in the EUR/USD

currency market on an intraday basis. The intraday timeframe of the Forex market had not been

extensively studied by researchers. Given the considerable turnover volume in the Forex market

and the need for currency traders to increase the net gains from trading activity, it is important to

49
further understand the application of technical analysis to intraday currency movements. The

results of this study might show evidence that using technical analysis on an intraday basis for

currency traders could increase the net gains from trading activity.

Types of technical trading analysis. Various technical trading tools are commonly used

by currency traders as a means to increase net gains for their respective companies (Neely et al.,

2009; Schulmeister, 2006; Zarrabi et al., 2017). Technical analysis and technical trading rules

can be broken down into various categories, and each category has its own unique characteristics

(Zarrabi et al., 2017). Moving average, both simple moving averages and exponential moving

averages, are trading rules attempt to identify a trend and identify when the trend might come to

an end by examing how prices interact with the moving average line (Deng & Sakurai, 2013;

Menkhoff & Taylor, 2007; Ozturk et al., 2016; Raj, 2013; Sobreiro et al., 2016). Moving

average crossover rules examine the interaction of two or more moving average lines. The

moving average technical indicator requires two moving average calculations, one fast and one

slow (Coakley et al., 2016; Hsu et al., 2016; McMillan & Speight, 2012; Milionis &

Papanagiotou, 2013). An example of a moving average crossover might consist of a 50-period

simple moving average and a 200-period simple moving average. When the 50-period simple

moving average crosses below the 200 periods moving average, this might indicate a selling

opportunity. As with most technical indicators, there is a great deal of subjectivity, and specific

inputs (moving average timeframes) can vary from trader to trader.

Support and resistance technical analysis rules use prior price lows as support and prior

price highs as resistance (Hsu et al., 2016; Neely et al., 2014; Zarrabi et al., 2017). These areas

of support and resistance seem to be pricing levels that currency prices have a difficulty

penetrating. Similarly, channel trading creates support and resistance based upon prior price

50
movements, but the lower and upper bound tend to vary over time creating a channel (Neely et

al., 1997; Nazario et al., 2017; Poti et al., 2014). Traders, for example, might sell a currency as

prices reach the upper bounds of the channel and buy when prices reach the lower bounds.

Filter rules used in technical analysis produce trade signals in the direction of the

prevailing trend when the currency price as moved a specified percentage above or below the

opening price (Hsu et al., 2016; Popovic & Durovic, 2014; Zarrabi et al., 2017). Oscillator

trading rules attempt to indicate situations when the currency is over-bought (prices are too high)

and oversold (prices are too low; Ni et al., 2015; Pukthuanthong et al., 2007). Oscillator analysis

does not seek to follow the trend, but instead looks for contrarian trades and mean reversion

(Metghalchi et al., 2014; Savin et al., 2007; Schulmeister, 2006). One of the technical analysis

tools that are mean reverting is the Bollinger Bands. This study focused on the use of Bollinger

Bands on an intraday timeframe.

Variables in the study. The Bollinger Bands technical trading indicator was designed

by John Bollinger and is based upon mean reversion and relative price (Coakley et al., 2016;

Temnov, 2017; Yan, Zhang, Lv, & Li, 2017). Bollinger Bands are a technical analysis tool that

is built upon the mean reverting properties of financial assets (Bollinger, 2002). Bollinger Bands

are designed to identify potential reversal patterns in currency pairs (Bajgrowicz & Scaillet,

2012; Hayes et al., 2016). The Bollinger Bands consist of three lines drawn on the price chart

for various currency and can be used across several financial instruments. The top line, or band,

of the Bollinger Bands, is typically set to a specified number of standard deviations above the

moving average line (da Costa, Nazário, Bergo, Sobreiro, & Kimura, 2015; Girma & Paulson,

1998; Kolková & Lenertová, 2016). The moving average line is the middle of the three lines

(Bollinger, 2002). The moving average is usually an exponential moving average, but can also

51
be set to a simple moving average. The lower line, or bottom band, of the Bollinger Bands, is

typically set to a specific number of standard deviations below the middle moving average line

(Bollinger, 2002; Ozturk et al., 2016; Velez & Capra, 2000). The upper and lower lines form a

band around the price movement of the currencies. These bands shift proportionately to

maintain a relatively high band and a relatively low band instead of an absolute high and low

band (Azizan & M’ng, 2010; Temnov, 2017; Xu & Yang, 2013; Yan et al., 2017). When

currency prices reach a high band level, this would indicate prices are currently overbought or

too high and present an opportunity to sell the currency or take a short position. Additionally,

when currency prices reach the lower band, this would indicate that currency prices are oversold

or too low compared to relative prices and present a buying opportunity (Chen et al., 2014). The

bands act as a signal for the beginning and end of price trends and price reversals.

Bollinger Bands have been studied in various financial markets and were found to be

profitable in the commodities futures energy market (Lubnau & Todorova, 2015; Girma &

Paulson, 1998, 1999). Bollinger Bands have also been applied to the equity futures market and

shown to produce statistically significant profits (Kathy, 2015; Yan et al., 2017). Researchers

applying Bollinger Bands to equity markets were able to capture unexpected price movement.

However, researchers ultimately concluded that moving averages tended to be more profitable

(Joseph & Terence, 2003; Williams, 2006) than Bollinger Bands. Coakley et al. (2016) studied

Bollinger Bands using daily currency prices for 22 currencies traded against the U.S. dollar and

found that the strategy produced an average annual return of 20.6%, which exceeded the

traditional buy and hold returns. Additional studies have looked at Bollinger Bands and the

Forex market on daily timeframes or longer (Bitvai & Cohn, 2015; Chen et al., 2014; Kolková &

52
Lenertová, 2016) and found that Bollinger Bands help to increase net gains from trading activity

in comparison to the buy and hold strategy.

Despite the extensive research into the use of technical analysis in the Forex market,

Bollinger Bands research on intraday or tick-level price movements is extremely limited (Savin

et al., 2007; Zarrabi et al., 2017). These time frames are essential to currency traders as they are

continually studying and analyzing market conditions throughout the day, not just the end of day

prices. Therefore, the examination of Bollinger Bands on an intraday currency price might have

positive implications for the currency traders and currency managers.

Buy and hold strategy. The buy and hold strategy is an investment strategy where an

asset is purchased with the belief that the value of the asset will increase over time creating a net

gain (Cohen & Cabiri, 2015). From the literature reviewed above, the buy and hold strategy is

often used as a base line to compare any alternative trading strategy. Technical analysis might

produce positive net gains, but those net gains must exceed the net gains realized from the buy

and hold strategy (Coe & Laosethakul, 2010; Hudson et al., 1996; Zhu et al., 2015).

Additionally, the study must incorporate transaction costs as technical analysis will generate

more trading costs than a buy and hold strategy (Ellis & Parbery, 2005; Frömmel & Lampaert,

2016; Zakamulin, 2014). Therefore, profits from technical analysis must be higher than the buy

and hold strategy after accounting for transaction costs.

Sell and hold strategy. A sell and hold strategy is an investment strategy where an asset

is sold to the market at a high price with the belief that the value of the asset will decrease over

time and can be bought back at a lower price creating a net gain (Mohamad, 2016). Shorting,

also referred to as selling, the market would be the contra-position to the buy and hold strategy

(Kelley & Tetlock, 2013; Lee, 2016). Shorting within the foreign currency market can also

53
represent a hedging position for traders at international banks and global hedge funds

(Hasbrouck, 2002; Cohen, Diether, & Mallory, 2007; Tornell & Yuan, 2012). Currency market

shorting is an important aspect in the price discovery process (Lee, 2016; Jorge & Augusto,

2016; Mohamad, 2016) and therefore has been included as a variable of interest in many

empirical studies (Alexander, 2000; Chen, Gau, & Liao, 2016; Gilje & Taillard, 2017;

Nwachukwa & Shitta, 2015). Therefore, profits from technical analysis must be higher than the

short and hold strategy after accounting for transaction costs.

Summary of the literature review. The focus of this study is to explore the relationship

between net gains and alternative trading techniques. Currency managers, working at hedge

funds or international financial institutions, can utilize these techniques and potentially increase

the net gains from their currency trading activity. In this literature review I discussed the

significant components and structure of the current foreign exchange market to ensure an

operational understanding of currency markets. Next, the following section of this research

study reviewed the two major market theories. Support and challenges for the EMH and the

AMH can be found in scholarly literature. The combination of the efficient market hypothesis,

specifically the weak-form, and the adaptive market hypothesis, specifically technical analysis,

provides the theoretical support for this study. Next, information in the literature review

compared and contrasted fundamental analysis and technical analysis. Various scholarly

findings supported the use of technical analysis in the Forex market as a means of predicting

short-term price movements and providing traders with a tool for increasing net gains. The

literature indicated that research of technical analysis must account for the returns that could

have been earned from a buy and hold strategy and account for transaction costs. Previous

academic researchers provided information that technical trading strategies, specifically

54
Bollinger Bands, applied to the Forex market might increase the positive net gains from trading

on an intraday basis.

Transition and Summary of Section 1

Understanding the movement of currency prices is a critical component for traders and

currency managers (Costantini, Cuaresma, & Hlouskova, 2016; Chen et al., 2014; El Ouadghiri

& Uctum, 2016; Mcmillan & Speight, 2012). Accurately forecasting exchange rates continues to

be an elusive endeavor despite the vast amount of academic and non-academic resources devoted

to the task (Dal Bianco et al., 2012). Currency managers working for global financial institutions

and international hedge funds should use every method at their disposal to increase the net gains

from foreign currency transactions as a means to maximize shareholder value (Morck, 2014).

55
Section 2: The Project

The weak form of the EMH states that historical prices are ineffective at increasing the

net gains from trading activity for currency traders because all historical price information is

known by all currency traders (Fama, 1970). Based upon the EMH technical analysis, which

depends on historical prices, should provide zero benefits to currency traders in an efficient

market. Due to the unique nature of the currency market, which is a decentralized over-the-

counter market, information regarding prices may not disseminate uniformly or efficiently

(Dowell-Jones, 2012; Metghalchi et al., 2014; Nurunnabi, 2012) creating opportunities for

currency traders to increase net gains using technical analysis.

The focus of this quantitative study is to explore the potential relationship between net

gains from currency trading and alternative trading techniques using technical analysis that could

be utilized by currency managers at hedge funds and international financial institutions. The

previous literature review provided an analysis of existing research in the area of currency

trading and alternative trading strategies. Almost all scholarly research has focused on daily

prices or longer timeframes with very few researchers focusing on intraday timeframes. The

intraday nature of data will be a distinction of this study as compared to previous research.

This section of the research study includes a discussion of the activities necessary to

perform the research, collect the data, and analyze the data. Included within this section is a

review of the purpose of this research project, data collection methods, selection of the data,

research methodology and research design, data analysis, and a review of validity and reliability

of the data. These activities were necessary to properly conduct the research based on the

problem statement and purpose statement.

56
Purpose Statement

The purpose of this quantitative study is to examine the net gains from alternative trading

techniques that can be utilized by currency managers working for international banks and hedge

funds when trading the EUR/USD currency on an intraday basis. The results may provide

additional insights for currency managers with regards to the use of alternative trading strategies

on an intraday basis within the currency markets. The findings of the research may have a direct

application to the business problem of increasing the net gain from foreign currency transactions

undertaking by currency management teams and currency traders within global financial

institutions and hedge funds. Also, this examination will add to the current literature on

technical analysis and help fill in the gap with regards to intraday time frames.

Role of the Researcher

The researcher in this study served several important roles for this project. In order to

ensure validity and reliability of this study, it is imperative to discuss the actions the researcher

performed during data collection and testing (Rudestam & Newton, 2015). The researcher

determined which research method and research design would be most appropriate to answer the

research question for this study. The researcher gathered all necessary data required to perform

the correlation study of currency exchange rates and Bollinger Band trading on an intraday basis.

This correlational research study included data that were generated and collected without any

regard of the purpose for this research study. Data collected in this manner allowed no

opportunity for the researcher to bias the data. The researcher served as the primary analyst of

the data, which included performing the statistical tests and reviewing the results of the data.

Finally, the researcher provided an interpretation of the results from the statistical tests to

determine if the information addressed the research question and corresponding hypothesis.

57
Participants

No human participants or organizational representatives were used in this research study.

Archival data gathered from a third party were the primary source for all data used for statistical

tests. Therefore, no measures to gain access to individual participants and ensure the ethical

protection of participants was necessary.

Research Method and Design

There are three main research methods (quantitative, qualitative, and mixed method)

employed by researchers. Within the quantitative method, there are multiple research designs:

experimental, quasi-experimental, descriptive, and correlation. Selecting the appropriate

research design is important to ensure that the research question is addressed. Choosing a proper

research design will depend upon the problem to be addressed, the research questions, and the

nature of the data available (Creswell, 2014; Parylo, 2012). For this study, a quantitative

correlational study was selected as the appropriate method for addressing the research questions

and purpose of the study.

Discussion of method. Researchers using quantitative methods seek to exam various

hypotheses by testing the relationship between numerical variables (Creswell, 2014).

Quantitative researchers tend to gather objective, historical, numerical sample data and then

analyze these sample sets using statistical tools. Researchers then gain insights based on the

results of the statistical analysis (Stake, 2010; Yin, 2011). The quantitative research method

supports the nature of the data collected for this study. More importantly, since the purpose of

this research was to determine if there is any correlation between net gain within foreign

exchange transactions and various trading techniques (Creswell, 2014), the quantitative research

method was used for this study.

58
Qualitative research was not an appropriate research method for the problem identified

because analyzing personal experiences would not be appropriate to address the research

question. At the core of qualitative research is a method that has a “reliance on human

perceptions and understanding” (Stake, 2010, p. 11). Scholars using qualitative research often

attempt to explain a particular social behavior or way of thinking (Yin, 2011). Multiple sources

of data can be collected in the qualitative research approach to explore the specific context that

may help explain a particular observed behavior (Creswell, 2014). The researcher using a

qualitative study approach seeks to contribute insights from existing concepts or attempts to

develop new concepts (Yin, 2011). Based on the purpose of this study, qualitative research

methods would not be applicable.

The mixed method approach incorporates both qualitative and quantitative research

methods. Researchers employing a mixed method approach attempt to provide a more thorough

and complete understanding of the research problem by those employing this method (Stake,

2010). By combining both methods, the researcher hopes to gain greater clarity than either the

qualitative or the quantitative approach can provide on its own (Creswell, 2014). The mixed

method approach was not an appropriate research method for the problem identified in this

study. The qualitative portion of the mixed method would not be appropriate to address the

research questions.

Discussion of design. A quantitative correlational research design was selected for this

study to examine the relationship between net gain from currency trading and alternative trading

strategies using Bollinger Bands. Correlation research design typically uses sample data

gathered from a population to determine if there is any relationship between the variables being

tested (Cash et al., 2016). Additionally, within correlation studies, there is no attempt by the

59
researcher to control any variables used in the study (Meyers et al., 2016). Random sample data

for this study were gathered for the EUR/USD foreign currency pair starting in January 2009 and

ending December 2016. No attempt to control any of the variables in the sample set was made.

The dependent variable, the net gain from trading, was documented by several other researchers

(Menkhoff & Taylor, 2007; Metghalchi et al., 2014; Bitvai & Cohn, 2015). Three different

trading techniques were the independent variables (Bollinger Bands [X1], buy and hold-long

strategy [X2], and sell and hold-short strategy [X3]). These techniques will be applied to the

data to determine if a specific trading technique has a higher positive correlation with net gains

and to determine if there is a statistical difference between the different group means. Trading

techniques used to increase net gains can be utilized by currency managers working for

international banks and hedge funds. Given the nature of the data and problem statement,

conducting a correlation study was justified as the research method for this study.

The research for this study was designed to answer the following research question: Do

currency trading managers at international banks and hedge funds experience a difference in net

gain from currency transactions when employing a traditional trading approach as compared to

an alternative trading strategy when trading the EUR/USD spot market on an intraday period?

For this study, net gains from trading was the dependent variable. Net gains as a dependent

variable are documented by other researchers (Menkhoff & Taylor, 2007; Metghalchi et al.,

2014; Bitvai & Cohn, 2015). Based on the research question, the first independent variable was

the alternative trading strategy using Bollinger Bands (X1). Based on the research question, the

second independent variable was the traditional trading technique using a buy and hold strategy

(X2). Based on the research question, the third independent variable was the traditional trading

technique using a sell and hold strategy (X3). These variables relate directly to the specific

60
problem addressed in this study, which is some managers of currency traders are unclear if using

alternative trading strategies on an intraday basis can increase the net gains from currency

trading.

Summary of Research Method and Design

In summary, the research for this study was designed to answer the following research

question: Do currency trading managers at international banks and hedge funds experience a

difference in net gain from currency transactions when employing a traditional trading approach

as compared to an alternative trading strategy when trading the EUR/USD spot market on an

intraday period? A quantitative correlational study was selected for this study to examine the

relationship between net gain from currency trading and alternative trading strategies using

Bollinger Bands to address the research question. Net gains from trading will be the dependent

variable. The traditional trading strategy of buy and hold, a strategy of shorting the EUR/USD,

and the alternative trading strategy will be the independent variables. Net gains as a dependent

variable are documented by other researchers. These variables relate directly to the specific

problem to be addressed in this study, which is some managers of currency traders are unclear if

using alternative trading strategies on an intraday basis can increase the net gains from currency

trading.

Population and Sampling

Foreign currencies are traded globally over-the-counter and are not governed by any

centralized exchange (Ackerman, 2016; El Ouadghiri & Uctum, 2016). The population of

currency exchange rates includes thousands of currency pairs such as EUR/USD, AUD/USD,

USD/JPY, and USD/CHF (Harvey, 2013). For this study, a sample was drawn from the

population of EUR/USD exchange rates. According to the Bank of International Settlement, the

61
USD and the EUR account for the largest portion of exchange turnover volume in the spot

market (April 2016). The EUR/USD was chosen because of the significant volume that these

two currencies represent in the overall Forex market. Several recent studies have also examined

the EUR/USD currency pair because of the significance of trading volume (Bush & Stephens,

2016; Costantini et al., 2016; Hamzaoui & Regaieg, 2016; Kolkova, 2017; Wong & Heaney,

2017). The sample was gathered for an eight-year period beginning January 2009 through

December 2016. This period was selected based on the availability of the data. Forex market is

traded 24 hours a day, five days a week. Weekends were excluded from sampling because the

FX market is closed. This time period produced an estimated 500,000 price changes per day

with a projected total number of prices observed reaching 70,000,000 over the sample period.

Using Microsoft Excel, a stratified random sample of 20 days were selected from each of

the eight years providing a sample of 160 observations (n=160). The sample size of 160 was

selected based upon the confidence level of 90% and a confidence interval of 10%. A 90%

confidence level and lower has been employed by other researchers testing various trading

techniques in the financial markets. Hsu et al. (2016) set the 90% confidence interval as their

threshold when testing technical trading tools in the foreign exchange market. Kim, An, and

Kim (2015) examined exchange rates and capital flow using a confidence level of 84%. Kraple

and O’Brien (2105) and Krapl (2017) studied the Forex cash flow and equity exposure of U.S.

multinational corporations using the 90% confidence level. Marshall, Musayev, Pinto, and Tang

(2012) studied the foreign exchange volatility and the impact of news announcements using both

the .10 level and .05 level. Gloede and Menkhoff (2014) examined the financial professional’s

ability to accurately forecast their own financial performance using the 90% confidence interval.

Finally, Crowder (2014), examined exchange rate equilibrium using purchasing power parity for

62
roughly 12 different currencies and various time horizons using the 90% confidence interval.

Based on these previous research parameters and the similarity to this study, the 90% confidence

level and 10% confidence interval level were selected for this study.

Data Collection

The focus of the data is on the Forex spot market or cash market. The FX spot market is

the currency market for immediate delivery with all trades settling within two days of the

transaction (Powers, 2016; Rzepczynski, 2008). The spot market does not include trades of

financial instruments but instead is an exchange of one currency for another at the current

exchange rate or spot rate (Gerber, 2016; Kamau et al., 2015). In contrast, the futures market

includes the same current pairs, but the exchange will take place on some date in the future

(Pintar et al., 2016). The spot market was chosen because this is the market that most global

bank traders and hedge fund traders participate (Wang et al., 2017). The distinction between the

spot market and futures is important because the spot market is decentralized and no single

source of data encapsulates all currency activity. The decentralization issue is overcome by

using a third party data aggregator as described in the next section.

Instruments. This quantitative research did not employ any specific instrument such as

questionnaires, interviews, or surveys. The data needed to conduct the statistical tests were

obtained from a third-party provider. The data were downloaded directly from the data

provider’s website in comma delimited format. This information was loaded directly into

Microsoft Excel. A Microsoft Excel template was created for this project to handle the creation

of Bollinger Bands, simple moving average, trade entry, trade exit, and calculate cumulative net

gains from each trading strategy. These results were then imported into SPSS to perform the

statistical analysis.

63
Data collection techniques. The Forex market is a globally decentralized over-the-

counter market (Ackerman, 2016; El Ouadghiri & Uctum, 2016). This means there were no

single source for all currency data transactions. Data for this study were obtained from a third

party data provider, TickData, LLC. TickData, LLC overcomes the issues of decentralization

and over-the-counter execution by aggregating currency quotes from several contributing

institutions. Each contributor provides primary, secondary, and tertiary data (TickData, LLC.,

2016). This process provides data that are isolated, redundant, and geographically diverse.

Through a proprietary data system, TickData, LLC compares all three levels of currency

exchange prices and synthesizes all information into a single data item free from errors and

omissions (TickData, 2017).

The information provided by Tickdata, LLC includes quote date, quote time down to the

millisecond, bid price, ask price, contributor code (source of exchange data), region code (region

of the exchange data), and city code (city of the exchange data). Tick-level data includes the

date, a time stamp, bid price, and ask price. The regional data are provided from North America

(NAM), Europe, the Middle East and Africa (EMEA), Asia-Pacific (ASI), and other global

locations (GLO).

Data organization techniques. Data for this study were organized using Microsoft

Excel. Currency tick data were collected for each random day selected for this study. Several

days in the study required more than one Excel file because there were more lines of data than

one Excel sheet could hold. Each day’s data were then processed for the opening bid price,

opening ask price, closing ask price, and closing bid price. Additionally, each day was processed

to obtain the net gain from each trading strategy.

64
Summary of Data Collection

In summary, the decentralization issue was overcome by using a third party data

aggregator, Tickdata, LLC. The data were downloaded directly from the data provider’s website

in comma delimited format. This information was loaded directly into Microsoft Excel. A

Microsoft Excel template was created for this project to handle the creation of Bollinger Bands,

simple moving average, trade entry, trade exit, and calculate cumulative net gains from each

trading strategy.

Data Analysis

Data obtained from Tickdata, LLC were directly exported into Microsoft Excel.

Microsoft Excel was used to establish the Bollinger Bands, trade entry, trade exit, calculate net

gains, and summation of trading activity. All Microsoft Excel files were stored in a cloud-based

storage system for easy access and retrieval. The cloud storage provider, DropBox, offers

password-protected access. However, because the data did not contain any sensitive

information, no additional security steps were taken to safeguard the data. The Microsoft Excel

data were then imported into IBM SPSS to perform statistical analysis of the data.

Variables used in the study. In order to address the research question for this study, the

first independent variable was the alternative trading strategy using Bollinger Bands (X1). Based

on the research question, the second independent variable was the traditional trading technique

using a buy and hold strategy (X2). Based on the research question, the third independent

variable was the traditional trading technique using a sell and hold strategy (X3). These

variables relate directly to the specific problem to be addressed in this study, which is some

managers of currency traders are unclear if using alternative trading strategies on an intraday

basis can increase the net gains from currency trading. The following paragraphs provide: (a) a

65
detailed explanation of all the variables used in this study, (b) a discussion of how those

variables will be collected, and (c) a review of how those variables will be tested.

Bollinger Band Parameters. Bollinger Bands are a technical analysis tool that is built

upon the mean-reverting properties of financial assets (Bollinger, 2002). Bollinger Bands are

designed to identify potential reversal patterns in currency pairs (Bajgrowicz & Scaillet, 2012;

Hayes et al., 2016). The Bollinger Bands consist of three lines drawn on the price chart for

various currency and can be used across several financial instruments. The top line, or band, of

the Bollinger Bands, is typically set to a specified number of standard deviations above the

moving average line (da Costa et al., 2015; Girma & Paulson, 1998; Kolková & Lenertová,

2016). The moving average line is the middle of the three lines (Bollinger, 2002). The moving

average can be a simple moving average or an exponential moving average. For this study, the

moving average is a simple moving average similar to other studies (Hayes et al., 2016, Kolková

& Lenertová, 2016).

For this study, the moving average was a simple moving average calculated using the

previous 20 price ticks of the mid-price. The mid-price was calculated as follows:

𝑏𝑏+𝑎𝑎
Mid-price =
2

Where b is the current bid price, and a is the current ask price. The simple moving

average will be calculated using the mid-price from the previous 20 price ticks. The simple

moving average was calculated as follows:

∑𝑛𝑛
𝑖𝑖=1 𝑃𝑃(𝑑𝑑−𝑖𝑖)+1
Simple Moving Average = 𝑛𝑛
𝑛𝑛 ≤ 𝑑𝑑

Where P is the mid-price, n is the number of periods, d is moving average periods. The

lower line, or bottom band, of the Bollinger Bands, is typically set to a specific number of

66
moving standard deviations below the middle moving average line. Similarly, the upper line, or

upper band, of the Bollinger Bands, is typically set to a specific number of moving standard

deviations above the middle moving average line (Bollinger, 2002; Ozturk et al., 2016; Velez &

Capra, 2000). The simple moving average and the moving standard deviation will both be

calculated on the same number of periods. For this study, the upper and lower bands were set at

3 standard deviations from the simple moving average over the previous 20 periods. The

standard deviation for each exchange rate for the previous 20 periods was calculated as follows:

∑(𝑥𝑥−𝑥𝑥)2
Moving Standard Deviation =�
𝑁𝑁

Where x is mid-price,  is the mean of the mid-price over the last 20 periods, and N is the

number of periods. The moving standard deviation formula does not use the sample standard

deviation formula because there is no attempt, at this point in the calculation, to infer sample data

onto population data (Lubnau & Todorova, 2015; Ozturk et al., 2016). Therefore, there is no

need for degrees of freedom adjustment that is contained in the sample standard deviation

calculation.

Bollinger Band Entry and Exit. Trade signals for entering and exiting positions are

described below. Entry signals for long and short positions for Bollinger Bands are generated

when EUR/USD prices reach extreme values (Lubnau & Todorova, 2015; Ozturk et al., 2016).

For this study, extreme values are defined as 3 standard deviations above the simple moving

average and -3 standard deviations below the simple moving average. Entry signals for long

currency trades occur whenever prices have reached an extreme value (-3 standard deviations

below the simple moving average), and then prices begin reverting toward the simple moving

average. A long position will be entered whenever prices cross back above the -3.0 standard

67
deviation line. Long positions will be executed at the current ask price. Long position profit

targets will be set at 50% of 1% of the current exchange rate for that trade. Long position stop

loss targets will be set at 50% of 1% of the current exchange rate for that trade.

Long positions will be closed at the bid price. This effectively captures the bid/ask

spread which would account for the transaction costs of the Bollinger Band strategy. The

difference of the bid/ask spread will effectively lower each net gain and increase each net loss as

would be experienced by real-time traders. Additionally, once a long trade has been entered into,

this position must be closed before any new trades can be executed (there will only be one trade

active at any one time regardless of the signals). Net gains from this strategy, inclusive of

transaction cost, will be combined with the net gains from the short Bollinger Band trades to

arrive at one figure for net gains from the Bollinger band strategy for that particular trading day.

Short positions will be entered into at the current bid price. Entry signals for short

currency trades occur whenever the exchange rate has reached an extreme value (a standard

deviation of 3), and then the exchange rate begins reverting toward the simple moving average.

A short position will be entered whenever the exchange rate crosses below the 3 standard

deviations after prices have traded above the 3 standard deviation value. Short positions will be

executed at the ask price. Short position profit targets will be set at 50% of 1% of the current

exchange rate for that trade. Short position stop loss targets will be set at 50% of 1% of the

current exchange rate for that trade price.

Short positions will be closed at the ask price. This effectively captures the bid/ask

spread which would account for the transaction costs of the Bollinger Band strategy. The

difference of the bid/ask spread will effectively lower each net gain and increase each net loss as

would be experienced by real-time traders. Additionally, once a long trade has been entered into,

68
this position must be closed before any new trades can be executed (there will only be one trade

active at any one time regardless of the signals). Net gains from this strategy, inclusive of

transaction cost, will be combined with the net gains from the short Bollinger Band trades to

arrive at one figure for net gains from the Bollinger Band strategy for that specific trading day.

Buy and hold: Parameters and entry/exit. In order to address the research question for

this study and the second hypothesis, a traditional buy and hold strategy will be examined.

Trades for the traditional buy and hold technique will be entered into at the beginning of the

trading day. Every trading day, a long currency trade will be opened at the beginning of the

trading day at the opening ask price for that day and closed at the end of the trading day at the

closing bid price for that day. These trades will represent a traditional trading strategy taking a

long position in the EUR/USD. Long transactions will be executed at the ask price and closed at

the bid price. The net gains from this strategy, inclusive of transaction costs equal to the bid/ask

spread, will be calculated for the sample period (Kuang et al., 2014; Lubnau & Todorova, 2015).

Sell and hold: Parameters and entry/exit. Simulating a short currency trade was

accomplished by opening a short currency position every day and closing that short position at

the close of every trading day. Short transactions were executed at the bid price and closed at the

ask price. Net gains from this strategy, inclusive of transactions costs equal to the bid/ask

spread, will be calculated for the sample period (Kuang et al., 2014; Lubnau & Todorova, 2015).

These net gains or losses for each trading day were then used to compare the variance between

the Bollinger Band strategy, buy and hold strategy, and the short and hold strategy.

Transaction size, transaction costs, and interest. The specific problem to be addressed

in this study was that some managers of currency traders are unclear if using alternative trading

strategies on an intraday basis can increase the net gains from currency trading. It is necessary to

69
replicate trading activities that are similar to real-world situations. The contract size for each

trade will be the standard lot size of 100,000 units of the base currency which is USD (Mende &

Menkhoff, 2006; Ozturk et al., 2016). Transaction costs will be calculated at the time of each

trade. Transaction costs will be equal to the difference between the bid/ask spread at the time of

the transaction (Levich & Thomas, 1993; Opie & Dark, 2015). This costs will be deducted from

the gain or added to any loss incurred by both the traditional trading strategy of buy and hold or

the alternative trading strategy using Bollinger Bands. For example, a long position in the

EUR/USD is entered into when the bid is 1.25603, and the ask is 1.25601. The transaction cost

would be equal to the difference between the bid and the ask price (0.00002) multiplied by

$100,000 for a total cost of $2.00 (Neely et al., 2009; Ozturk et al., 2016). Upon exiting the

position, the same method to calculate the transaction costs will be used. Each trade will have

two transactions costs, one to open the trade and one to close the trade.

Given that all trading began and ended within the same trading day, the researcher

assumed that traders would not gain any interest for holding currencies overnight. Additionally,

no interest is earned for holding cash positions (Dunis & Evans, 2006; Lubnau & Todorova,

2015; Metghalchi et al., 2014; Narayan et al., 2015).

Hypotheses 1. The stated hypotheses directly relate to the research question and attempt

to address the specific problem of this research study. The first null hypothesis and

corresponding alternative hypothesis examines the relationship between the alternative trading

strategy using Bollinger Bands and the traditional buy and hold strategy. It states:

H10: There is no statistically significant difference between net gains generated and the

use of an alternative trading strategy (X1) on an intraday basis for the EUR/USD spot currency

market as compared to the traditional buy and hold strategy (X2).

70
H1a: There is a statistically significant difference between net gains generated and the

use of an alternative trading strategy (X1) on an intraday basis for the EUR/USD spot currency

market as compared to the traditional buy and hold strategy (X2).

Hypotheses 2. The second null hypothesis and corresponding alternative hypothesis

examines the relationship between the alternative trading strategy using Bollinger Bands and the

traditional sell and hold strategy. It states:

H20: There is no statistically significant difference between net gains generated and the

use of an alternative trading strategy (X1) on an intraday basis for the EUR/USD spot currency

market as compared to the sell and hold strategy (X3).

H2a: There is a statistically significant difference between net gains generated and the

use of an alternative trading strategy (X1) on an intraday basis for the EUR/USD spot currency

market as compared to the sell and hold strategy (X3).

Statistical analysis. The one-way analysis of variance (ANOVA) was used to test for

significant differences between the independent variables (treatments). The dependent variable

in this study is the net gains from trading activity. The treatments or independent variables are

the Bollinger band strategy (X1), buy and hold strategy (X2), and short and hold strategy (X3).

As described above Microsoft Excel was used to generate a random sample of 20 days per year

between January 2009 and December 2016. Net gains (dependent variable) for each independent

variable was determined using the same tick level currency data. Therefore, the only difference

between the independent variables are the trading strategies.

In order to generate a net gain figure from the Bollinger Band strategy (X1), the raw tick

level data are imported into Microsoft Excel, and the trading parameters described above

(Bollinger Band Entry and Exit) were applied to the data. This procedure produced a set of

71
trades using the Bollinger Band strategy. The gains and losses for all trades from that specific

trading day were aggregated together to arrive at a single net gain or net loss from the Bollinger

Band strategy for each randomly selected day. The net gains from this strategy were used to test

Hypothesis 1 and Hypothesis 2.

In order to generate a net gain figure (dependent variable) for each random day for the

buy and hold strategy (X2), a long currency trade was opened at the beginning of the trading day

at the opening ask price for that day and closed at the end of the trading day at the closing bid

price for that day. These trades represented a traditional buy and hold trading strategy taking a

long position in the EUR/USD. Long transactions were executed at the bid price and closed at

the ask price. The net gains from this strategy, inclusive of transaction costs equal to the bid/ask

spread, were calculated for the sample period and entered into Microsoft Excel for that day. The

net gains from this strategy were used to test Hypothesis 1.

In order to generate a net gain figure (dependent variable) for each random day for the

sell and hold strategy (X3), a short currency trade was accomplished by opening a short currency

position every day and closing that short position at the close of every trading day. Short

transactions were executed at the ask price and closed at the bid price. The net gains from this

strategy, inclusive of transaction costs equal to the bid/ask spread, were calculated for the sample

period and entered into Microsoft Excel for that day. The net gains from this strategy were used

to test Hypothesis 2.

These net gains or losses for each trading day was then used to compare the average net

gains of the Bollinger Band strategy, the buy and hold strategy, and the short and hold strategy

using ANOVA. The level of significance for this study was alpha = .10. The 0.10 level of alpha

72
provides a 90% confidence level that the null hypothesis will be rejected when it is false.

Support for these levels were previously cited.

To conclude, NumXL, an add-on to Microsoft Excel, was used to test for normality of the

data. If the data are found to be normally distributed, then Microsoft Excel was used to perform

the ANOVA tests. Microsoft Excel was used to perform any additional post-hoc testing if

necessary.

Summary of data analysis. To address the research question for this study and the

related hypotheses, an alternative trading technique using Bollinger Bands was examined. Data

obtained from Tickdata, LLC were directly exported into Microsoft Excel. Microsoft Excel was

used to establish the Bollinger Bands, trade entry, trade exit, calculate net gains, and summation

of trading activity. Trades for the traditional buy and hold technique were entered into at the

beginning of the trading day. Every trading day, a long currency trade was opened at the

beginning of the trading day at the opening ask price for that day and closed at the end of the

trading day at the closing bid price for that day. Short transactions were executed at the ask price

and closed at the bid price. Net gains from this strategy, inclusive of transactions costs equal to

the bid/ask spread, were calculated for the sample period. These net gains or losses for each

trading day were then used to compare the variance between the Bollinger Band strategy, buy

and hold strategy, and the short and hold strategy.

Reliability and Validity

Reliability and validity are essential in all scholarly research (Creswell, 2014).

Reliability addresses the replicability of the study under similar circumstances and attempts to

understand any measurement error that might exist (Muijs, 2011). The concept of validity seeks

to ensure that the research tools being used are measuring what they are supposed to and directly

73
relate back to the hypothesis (Creswell, 2014). Reliability and validity need to be present in

research studies so the findings and conclusions can be examined by other researchers.

Reliability. Reliability in research examines the issue of replication of the study under

similar circumstances (Fink, 2014). The application of reliability varied depending on the

research method (qualitative, quantitative, or mixed method). In qualitative studies, reliability

can address issues of training interviewers, describing a systematic approach to recording data,

and theme development (Yin, 2006). Reliability in qualitative studies focuses on consistency in

coding raw data in such a way that other researchers could arrive at similar conclusions

(Rudestam & Newton, 2015). In quantitative research methods, reliability addresses the ability

of other researchers to confirm the results of the study and examines the accuracy of the data

collection process (Creswell, 2014). Given this study used a quantitative research approach, the

focus of reliability was on replication of the results and understanding the accuracy of the data.

The researcher used archival data for this research project, and no other testing

instrument was used. Archival data were provided by TickData, LLC, a third party data provider

for currency market data. Historical market data provided by TickData have been used in several

other market research studies which have appeared recently in scholarly peer-reviewed journals

(Bastidon, 2017; Drechsler, 2013; Ishida, McAleer, & Oya, 2011; Jubinski & Lipton, 2012;

Lasser & Spizman, 2016; Prokopczuk, Symeonidis, & Wese-Simen, 2016; Wright, 2012). Usage

of the archival currency data provided by TickData by other researchers supports the reliability

of this study. However, there is an inherent data issue with all historical spot currency market

data. The foreign currency exchange market is a decentralized over-the-counter market

(Ackerman, 2016; El Ouadghiri & Uctum, 2016). Therefore, one single source for all currency

data transactions does not exist. TickData, LLC attempts to overcome the issues of

74
decentralization and over-the-counter execution by aggregating currency quotes from several

contributing institutions. Each contributor provides primary, secondary, and tertiary data. This

process provides data that are isolated, redundant, and geographically diverse. Through a

proprietary data system, TickData, LLC compares all three levels of currency exchange prices

and synthesizes all information into a single data item free from errors and omissions (TickData,

2017). Based upon this process and the use of TickData information by other researchers, it is

reasonable to assume the currency data used in this study accurately represents the historical

currency prices of the EUR/USD spot exchange market. It is also reasonable to assume that

researchers using other data sources would reach similar conclusions.

Validity. Validity in the context of quantitative research “refers to the degree to which a

measure assesses what it purports to measure” (Fink, 2014, p. 106). The validity of research

studies can be broken down into the internal validity and the external validity. The researcher

discusses both internal and external validity issues below.

Internal validity. Internal validity refers to the ability of the study to claim a causal

relationship between independent variables and the dependent variable (Creswell, 2014;

Rudestam & Newton, 2015). In this quantitative correlation study, the researcher was not

attempting to claim a causal relationship between the independent variables (trading strategies)

and the dependent variable (net gains). Instead, the researcher sought to understand the

correlation of the relationship, the direction of the relationship, the strength of the relationship,

and the statistical significance of the relationship between the independent variables and

dependent variable. Additionally, the researcher used historical archival currency data provided

by an independent third party to reduce the threat to internal validity. There were no other data

collection methods used in this research study.

75
External validity. External validity of quantitative research methods addresses the issue

of generalizability of the sample data to make inferences about the population as a whole

(Creswell, 2014). Threats to external validity in research studies occur when generalizability of

the results do not transfer to other samples, times, or situations (Taylor, 2014). To avoid the

threat of external validity due to sample selection, the researcher focused on an entire year of

data from the EUR/USD exchange rates. No attempt was made to bias the data set toward a

favorable or non-favorable outcome for any of the independent variables selected. The

researcher could not overcome the threat to external validity regarding time. Since this study is

not a replication of previous research, the only way to overcome this threat is to replicate this

study over different time periods (Creswell, 2014). This threat will be addressed in the

recommendations for future research. The threat to external validity posed by the situation, or

setting, was minimized by using historical archival data from the same source for all tests and

using similar data processing.

Summary of reliability and validity. Reliability and validity need to be present in

research studies so that the findings and conclusions can be examined by other researchers.

Reliability addresses the replicability of the study under similar circumstances and attempts to

understand any measurement error that might exist. The concept of validity seeks to ensure that

the research tools being used are measuring what they are supposed to and directly relate back to

the hypothesis. Reliability and validity are essential for all scholarly research.

Transition and Summary of Section 2

The focus of this quantitative study was to explore the potential relationship between net

gains from currency trading and alternative trading techniques using technical analysis that could

be utilized by currency managers at hedge funds and international financial institutions. Almost

76
all scholarly research has focused on daily prices or longer timeframes with very few focusing on

intraday. The intraday nature of data will be a distinction of this study as compared to previous

research. This section reviewed the purpose of the study, the stated research question, and the

hypothesis formed to answer the research question. Additionally, this section discussed the

research methodology and quantitative design, data collection, data analysis, reliability, and

validity. Consideration was given to each step of this research study to maximize the validity

and reliability of the findings presented in this research study.

The next section of this research project will discuss in detail the results of the data

analysis, correlations, strength of the relationship, ANOVA results, and confidence in the data.

Due to the applied nature of the research problem, the next section will also include

recommendations for application in currency trading taking place at international banks and

hedge funds. Finally, recommendations for further study will be made to support the validity

and reliability of the findings of this study.

77
Section 3: Application to Professional Practice and Implications for Change

International financial institutions and hedge funds speculate in currency markets as a

means to increase profits for their existing currency positions (Aktan et al., 2013; Shen &

Hartarska, 2013). The role of currency traders, whether they are working for multinational

financial institutions or hedge funds, is to engage in currency activities that increase the net gain

per currency transaction for their respective firms (Fiedor & Holda, 2016; Kamau et al., 2015).

Currency traders use technical analysis, fundamental analysis, or a combination of both as a

means to increase net gains from trading activity (Kuang et al., 2014; de Zwart et al., 2009).

Technical analysis tools such as the Bollinger Bands can be used by currency traders and may

support increased net gains realized from their foreign exchange trading activity (Coakley et al.,

2016; Chen et al., 2014; Lubnau & Todorova, 2015). The purpose of this quantitative study is to

examine the net gains from alternative trading techniques that can be utilized by currency

managers working for international banks and hedge funds when trading the EUR/USD currency

on an intraday basis.

The findings from this research are presented in this section. These findings may

potentially contribute to the existing body of knowledge concerning the use of technical analysis

when trading the EUR/USD currency pair on an intraday basis. This section includes: (a)

overview of the study, (b) presentation of the findings, (c) applications to professional practice,

(d) recommendation for further action, (e) recommendation for further study, (f) reflections, and

(g) summary and conclusion of the study.

Overview of the Study

The challenge for many international firms is that essential currency markets are also a

source of volatility and risk for all participants (Wong, 2016). Global financial institutions

78
provide a mechanism for multinational corporations to hedge against exchange rate risk via

currency futures contracts and spot exchange rates (Chien et al., 2013). Within these global

financial institutions, the currency trading team, led by currency managers, speculate in currency

markets as a means to increase profits of their existing currency positions (Aktan et al., 2013;

Shen & Hartarska, 2013). This research project was undertaken to add to the current body of

knowledge concerning currency trading and technical analysis.

The research for this study was designed to answer the following research question: Do

currency trading managers at international banks and hedge funds experience a difference in net

gain from currency transactions when employing a traditional trading approach as compared to

an alternative trading strategy when trading the EUR/USD spot market on an intraday period?

The findings from this research suggest that using an alternative trading strategy, such as the

Bollinger Bands, on an intraday basis does not necessarily increase net gain from trading

activity. The net gain from trading activity using the Bollinger Bands on an intraday basis

produced an average negative return. This is similar to the results of the buy and hold strategy

which also produced an average negative return over the same sample period. However, the sell

and hold strategy produced a positive average return over the sample period. A detailed

discussion of the findings from this study is provided in the following section.

Presentation of the Findings

The data available for this research study included historical tick data for the EUR/USD

from 2009 to 2016. The data were purchased from TickData, LLC. A stratified random sample

of 20 days per year (excluding weekends) was selected producing a sample size of 160 days

(n=160). There were no gaps or missing data for the random days selected. Appendix A lists all

79
160 days that were randomly generated using Microsoft Excel’s random function. These days

are provided for those who wish to examine the specific trading days included in the study.

Descriptive Statistics

Table 1 below provides the descriptive statistics for the Bollinger Band trading strategy,

the buy and hold strategy, and the sell and hold strategy for all eight years.

Table 1

Descriptive Statistics for All Eight Years

Bollinger Band Strategy Buy and Hold Strategy Sell and Hold Strategy

Mean Net Gain $ (1,128.61) Mean Net Gain $ (65.81) Mean Net Gain $ 16.88
Standard Error 248.27 Standard Error 60.32 Standard Error 60.24
Median $ (70.00) Median $ 10.00 Median $ (60.00)
Standard Deviation $ 3,140.38 Standard Deviat $ 763.00 Standard Devia $ 761.97
Sample Variance 9,862,014.24 Sample Varianc 582,166.63 Sample Varian 580,601.49
Kurtosis 3.21 Kurtosis 3.31 Kurtosis 3.28
Skewness -1.74 Skewness -0.81 Skewness 0.78
Range $ 19,480.00 Range $ 5,500.00 Range $ 5,480.00
Minimum $ (13,740.00) Minimum $ (3,320.00) Minimum $ (2,250.00)
Maximum $ 5,740.00 Maximum $ 2,180.00 Maximum $ 3,230.00
Sum $ (180,578.00) Sum $ (10,530.00) Sum $ 2,700.00
Count 160 Count 160 Count 160

Table 1 displays the mean net gain from each trading strategy. Over the entire sample

period (n=160), the mean net gain from using the Bollinger Band strategy produced a negative

gain of $1,128.61 for each trading day with a total negative gain of $180,578. The mean net gain

from using the buy and hold strategy was a negative $65.81 per trading day with a total negative

gain of $10,530. The sell and hold strategy produced a mean net gain of $16.88 per trading day

with a total positive net gain of $2,700.

Table 2 displays the descriptive statistics on a year-by-year basis for each trading

strategy. From Table 2, the Bollinger Band strategy produced positive mean net gains in 2009,

2010, and 2014. For the Bollinger Band strategy, the highest positive net gain was in 2009 for a

80
total gain of $17,330, while 2015 showed the largest negative net loss of $132,426. The buy and

hold strategy produced positive mean net gains in 2012 and 2016. For the buy and hold strategy,

the highest positive net gain was in 2012 in the amount of $2,000 while the largest net loss was

in 2009 in the amount of $9,540. The sell and hold strategy produced positive mean net gains in

2009, 2010, and 2013. For the sell and hold strategy, the highest positive net gain was in 2009 in

the amount of $8,250 while the largest net loss was in 2012 in the amount of $3,100.

Table 2

Year-by-Year Descriptive Statistics for each Trading Strategy

Bollinger Bands
Mean Net Gain SE Median Std. Dev. Sample Var. Kurtosis Skewness Range Sum Sample size (n)
2009 $ 866.50 420.74 470 1881.60 3,540,434.47 1.08 1.09 $ 7,280.00 $ 17,330.00 20
2010 $ 106.50 151.99 95 679.72 462,013.42 -0.77 0.34 $ 2,450.00 $ 2,130.00 20
2011 $ (111.50) 172.46 35 771.27 594,855.53 -0.08 0.39 $ 2,910.00 $ (2,230.00) 20
2012 $ (25.00) 102.11 -35 456.65 208,531.58 1.01 -0.21 $ 2,060.00 $ (500.00) 20
2013 $ (135.50) 142.57 -25 637.61 406,541.84 1.48 -1.05 $ 2,710.00 $ (2,710.00) 20
2014 $ 32.40 53.95 25 241.29 58,219.94 -0.09 0.30 $ 880.00 $ 648.00 20
2015 $ (6,621.30) 798.42 -6660 3570.65 12,749,537.06 0.39 0.22 $ 14,380.00 $ (132,426.00) 20
2016 $ (3,141.00) 945.12 -2490 4226.72 17,865,146.32 -0.55 -0.02 $ 15,050.00 $ (62,820.00) 20

Buy and Hold

Mean Net Gain SE Median Std. Dev. Sample Var. Kurtosis Skewness Range Sum Sample size (n)
2009 $ (477.00) 297.09 -300.00 1328.63 1,765,264.21 0.62 -0.27 $ 5,370.00 $ (9,540.00) 20
2010 $ (91.50) 195.52 -125.00 874.39 764,550.26 1.04 0.83 $ 3,640.00 $ (1,830.00) 20
2011 $ (38.00) 161.47 160.00 722.10 521,427.37 1.08 -1.00 $ 2,910.00 $ (760.00) 20
2012 $ 100.00 127.88 85.00 571.91 327,084.21 -0.21 -0.27 $ 2,030.00 $ 2,000.00 20
2013 $ (55.50) 108.18 -30.00 483.80 234,057.63 1.05 -0.96 $ 1,810.00 $ (1,110.00) 20
2014 $ (20.50) 73.29 -20.00 327.74 107,415.53 0.91 0.75 $ 1,300.00 $ (410.00) 20
2015 $ (13.00) 169.71 30.00 758.96 576,022.11 3.70 -1.33 $ 3,480.00 $ (260.00) 20
2016 $ 69.00 131.00 105.00 585.86 343,230.53 -0.48 -0.37 $ 1,970.00 $ 1,380.00 20

Sell and Hold

Mean Net Gain SE Median Std. Dev. Sample Var. Kurtosis Skewness Range Sum Sample size (n)
2009 $ 412.50 297.39 225.00 1329.98 1,768,840.79 0.59 0.26 $ 5,350.00 $ 8,250.00 20
2010 $ 27.00 195.73 65.00 875.35 766,232.63 1.06 -0.83 $ 3,630.00 $ 540.00 20
2011 $ (19.50) 160.36 -200.00 717.15 514,310.26 1.01 0.97 $ 2,880.00 $ (390.00) 20
2012 $ (155.00) 128.80 -150.00 576.03 331,805.26 -0.16 0.31 $ 2,080.00 $ (3,100.00) 20
2013 $ 12.00 109.32 -50.00 488.88 239,006.32 1.18 1.01 $ 1,860.00 $ 240.00 20
2014 $ (15.00) 73.41 -5.00 328.32 107,794.74 0.74 -0.75 $ 1,270.00 $ (300.00) 20
2015 $ (23.00) 170.00 -80.00 760.28 578,032.63 3.70 1.34 $ 3,480.00 $ (460.00) 20
2016 $ (104.00) 129.84 -130.00 580.68 337,193.68 -0.40 0.32 $ 2,010.00 $ (2,080.00) 20

The positive mean net gains generated by the Bollinger Band strategy are consistent with

those of Coe and Laosethakul (2010), Hudson et al. (1996), and Zhu et al. (2015), who

discovered that alternative trading strategies using technical analysis can generate profitable

81
trades. However, the negative mean net gains generated by the Bollinger Band Strategy are

consistent with those of Anderson and Faff (2005), Bajgrowicz and Scaillet (2012), and Sullivan

et al. (1999), who found that technical tools like Bollinger Bands are not effective in generating

profits from trading. The positive net gains and the negative net gains support the accretions of

the AMH (Adaptive Market Hypothesis).

As previously discussed on the literature review, the main premise of the AMH is that

market efficiency is not a steady state, and therefore alternative trading rules can be employed to

achieve positive net gains (Lo, 2004). Based on the descriptive statistics presented in Table 2,

the mean net gains using the Bollinger Band strategy produces both positive and negative gains.

This is consistent with the non-constant efficient market premise found in the AMH. Another

premise of the AMH is that market participants will adapt in order to gain profits and therefore

profitable trading strategies will diminish over time (Noda, 2016). Referring to the descriptive

statistics in Table 2, it is unclear whether the data support this premise of the AMH. If

diminishing net gains were occurring, a Bollinger Band strategy that produced positive net gains

at the beginning of the sample period would be expected to have diminishing positive net gains

over the sample period. However, Table 2 shows the Bollinger Band strategy was profitable in

the earlier years of the sample period (2009 and 2010), then produced negative net gains (2011,

2012, and 2013), then produced positive net gains in 2014, and finally the Bollinger Bands

produced extremely negative net gains for 2015 and 2016. This does not correlate with the

diminishing returns phenomena as described by the adaptive market hypothesis.

Additional observations from Table 2 indicate that the Bollinger Band strategy produced

the largest net gains and largest net losses of any of the three trading strategies. The Bollinger

Band strategy has the largest range and standard deviation in five out of the eight years. The

82
largest range and standard deviation for the Bollinger Band strategy occured in 2015 and 2016

which correspond to the largest mean net losses for this strategy.

Normality

Prior to testing the hypotheses, the data were tested for normal distribution to ensure that

parametric testing would be valid. The following normality tests were performed for each of the

sample data using Microsoft Excel and an add-on data analysis tool provided by NumXL: the

Jarque-Bera, the Sharipo-Wilk, and the Doornik Chi-Squared. All three tests require a null

hypothesis stating that the population is normally distributed and the alternative hypothesis

stating that the population is not normally distributed. Using an alpha of 0.05, all three tests

provided a p-value for comparison. The lowest p-value, 0.1144, was obtained by the Doornik

Chi-Square test for the 2012 sample. Appendix 2 provides the results by year for each test using

an alpha of 0.05. The highest p-value, 0.8470, was obtained by the Shapiro-Wilk test for the

2010 sample. At the alpha of 0.05, all sample data for each year passed the test of normality.

These findings align with those of other researchers analyzing similar returns in currency trading

(Bush & Stephens, 2016; Caporale, Gil-Alana, & Plastun, 2018; Hsu et al., 2016; Kumar, 2016).

Based on the null hypothesis for all three tests is that the population is normally distributed, there

is not enough statistical evidence to reject the null hypothesis. Given that the sample data are

normal, the ANOVA test can be performed. A discussion of the hypotheses test is provided in

the next section.

Hypothesis 1

The research for this study was designed to answer the following research question: Do

currency trading managers at international banks and hedge funds experience a difference in net

gain from currency transactions when employing a traditional trading approach as compared to

83
an alternative trading strategy when trading the EUR/USD spot market on an intraday period?

This research question leads to the formation of the hypotheses for this study. Hypothesis 1

states:

H10: There is no statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the traditional buy and hold strategy.

H1a: There is a statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the traditional buy and hold strategy.

To address hypothesis 1, an ANOVA test was performed to determine if the difference

between the mean net gains for each of the trading strategies was statistically significant. Table

3 below shows the results of the ANOVA test. The p-value for the F-statistic is 0.00000000361.

This p-value is considerably smaller than the alpha of 0.10, indicating that there is a statistically

significant difference between all three trading strategies.

Table 3

ANOVA

SUMMARY
Groups Count Sum Average Variance
Bollinger Band Strategy 160 (180,578.00) (1,128.61) 9,862,014.24
Buy and Hold Strategy 160 (10,530.00) (65.81) 582,166.63
Sell and Hold Strategy 160 2,700.00 16.88 580,601.49

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 130,587,876 2 65,293,938 17.77 0.000000036 3.01463
Within Groups 1,752,940,396 477 3,674,927

Total 1,883,528,272 479

Based on the results of the ANOVA test and corresponding p-value for the F distribution,

there is a statistically significant difference between the three different trading strategies.

84
However, in order to reject the null hypothesis and accept the alternative hypothesis, as stated in

hypothesis 1, a post-hoc t-test must be performed to determine where the difference exists. The

t-test is used to determine if the statistically significant difference identified by the ANOVA test

is between the Bollinger Band strategy and the buy and hold strategy. The results of the t-test

can indicate a significant difference between the Bollinger Band strategy and the buy and hold

strategy. Table 4 below presents the results of the post-hoc t-test.

Table 4

t-Test Comparison of Bollinger Band Strategy and Buy and Hold Strategy

Bollinger Bands Buy and Hold

Mean (1,128.61) (65.81)
Variance 9,862,014.24 582,166.63
Observations 160.00 160.00
Pooled Variance 5,222,090.44
df 318.00
t Stat (4.16)
P-value one-tail 0.0000205
t Critical one-tail 1.65
P-value two-tail 0.0000410
t Critical two-tail 1.97

Table 4 displays a p-value of 0.0000410 for a two-tailed test of differences. This

indicates that there is a statistically significant difference between the Bollinger Bands and the

buy and hold strategy. Therefore, the null hypothesis for hypothesis 1 is rejected and the

alternative hypothesis is accepted. However, both the Bollinger Band strategy and the buy and

hold strategy produced a mean negative net gain for the sample period.

The p-value provided by the t-test results indicates that there is a statistical difference

between the Bollinger Bands and the buy and hold strategy. To determine the magnitude of the

difference a Cohen d-test for effect size was performed. The Cohen d-test provides an indication

as to the magnitude of the difference between to the two trading methods (Faul, Erdfelder, Lang,

85
& Buchner, 2007; Ferguson, 2009). The effect size d-statistic was calculated for the Bollinger

Bands and the sell and hold strategy as 0.46. Inputting the d-statistic into G*Power 3.1 software

the power or magnitude of the difference between these two methods was .9939. The power of

.9939 indicates the probability of rejecting the null hypothesis assuming that the alternative

hypothesis is true (Khalilzadeh & Tasci, 2017). According to Cohen (1988), any value over 0.80

for a t-test would indicate a large effect size. This additional information helps to add to the

generalizability of the sample to the overall population.

The Bollinger Band strategy demonstrated a mean negative net gain for the eight-year

sample period, indicating that the alternative trading strategy does not effectively increase the net

gain for currency traders on an intraday basis. Additionally, rejecting the null hypothesis

indicates the mean negative gains from the Bollinger Band strategy are significantly different

than the mean negative gains from the buy and hold strategy. These findings confer with the

weak-form of the EMH in that using historical prices does not allow currency traders to earn

excess profits over time. Additionally, these findings are consistent with the findings of other

researchers (Hsu et al., 2016; Westerlund & Narayan, 2013; Kofarbai & Zubairu, 2016) who

have examined other financial markets and various time-frames. The findings presented in this

study are unique in that I have examined the spot EUR/USD currency market at a tick-level.

The Bollinger Band strategy, as an alternative trading method, produced an overall mean

negative net gain that was significantly different than the mean negative net gains of the buy and

hold strategy. These findings are consistent with the weak-form of the EMH signifying that

historical pricing data is not effective in helping currency traders increase net gains from trading

activity. These results would indicate that the Bollinger Band strategy would produce larger

negative returns and should not be considered by currency trading managers for implementation.

86
Hypothesis 2

The research question produced a second hypothesis to be tested. Similar to hypothesis

one, hypothesis 2 examines the relationship between the Bollinger Band strategy and the sell and

hold strategy. Hypothesis two is stated as follows:

H20: There is no statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the sell and hold strategy.

H2a: There is a statistically significant difference between net gains generated and the

use of an alternative trading strategy on an intraday basis for the EUR/USD spot currency market

as compared to the sell and hold strategy.

To address hypothesis 2, an ANOVA test was performed to determine if the difference

between the mean net gains for each of the trading strategies was statistically significant. Table

3 shows the results of the ANOVA test. The p-value for the F-statistic is 0.00000000361. This

p-value is considerably smaller than the alpha of 0.10 indicating that there is a statistically

significant difference between all three trading strategies.

As previously mentioned, the results presented in the ANOVA test indicate that there is a

statistically significant difference between the three different trading strategies. However, in

order to reject the null hypothesis and accept the alternative hypothesis, as stated in hypothesis 2,

a post-hoc t-test must be performed. The t-test is used to determine if the statistically significant

difference identified by the ANOVA test is between the Bollinger Band strategy and the sell and

hold strategy. The results from the t-test, presented below, indicates that there is a significant

difference between the Bollinger Band strategy and the buy and hold strategy. Table 5 presents

the results of the post-hoc t-test.

87
Table 5

t-Test Comparison of Bollinger Band Strategy and Sell and Hold Strategy

Bollinger Bands Sell and Hold

Mean (1,128.61) 16.88
Variance 9,862,014.24 580,601.49
Observations 160.00 160.00
Pooled Variance 5,221,307.87
df 318.00
t Stat (4.48)
P-value one-tail 0.0000051
t Critical one-tail 1.65
P-value two-tail 0.0000103
t Critical two-tail 1.97

Table 5 displays a p-value of 0.0000103 for a two-tailed test of differences. This

indicates that there is a statistically significant difference between the Bollinger Bands and the

sell and hold strategy. Therefore, the null hypothesis for hypothesis 2 is rejected and the

alternative hypothesis is accepted. However, the Bollinger Bands produced a mean negative net

gain from trading. Based upon this finding, currency traders and currency managers should be

extremely cautious when using Bollinger Bands as a method to increase net gains from trading.

However, the sell and hold strategy produced a mean positive net gain from trading and therefore

should be considered by currency traders as a method to increase net gains from trading.

The p-value from the t-test indicates that there is a statistical difference between the

Bollinger Bands and the sell and hold strategy. To determine the magnitude of the difference a

Cohen d-test for effect size was performed. The Cohen d-test provides an indication as to the

magnitude of the difference between to the two trading methods (Faul et al., 2007; Ferguson,

2009). The effect size d-statistic was calculated for the Bollinger Bands and the sell and hold

strategy as 0.5031. Inputting the d-statistic into G*Power 3.1 software the power or magnitude

of the difference between these two methods was .9937. The power of .9973 indicates the

88
probability of rejecting the null hypothesis assuming that the alternative hypothesis is true

(Khalilzadeh & Tasci, 2017). According to Cohen (1988), any value over 0.80 for a t-test would

indicate a large effect size. This additional information helps to add to the generalizability of the

sample to the overall population.

The Bollinger Band strategy, representing alternative trading strategies, produced a mean

negative net gain for the eight-year sample period indicating that the alternative trading strategy

does not effectively increase the net gain for currency traders. Additionally, rejecting the null

hypothesis indicates the mean negative gains from the Bollinger Band strategy are significantly

different than the mean positive gains from the sell and hold strategy. These findings confer

with the weak-form of the EMH in that using historical prices does not allow currency traders to

earn excess profits over time. The sell and hold strategy is successful when the Euro weakens

against the U.S. dollar. Although this strategy is the inverse of the buy and hold strategy, the

findings would still be consistent with the findings of other researchers (Hsu et al., 2016;

Westerlund & Narayan, 2013; Kofarbai & Zubairu, 2016) who have examined other financial

markets and various time-frames using a buy and hold strategy.

The Bollinger Band strategy, as an example of an alternative trading method, produced

an overall negative return that was significantly different than the positive returns of the sell and

hold strategy. These findings are consistent with the weak-form of the EMH signifying that

historical pricing data is not effective in helping currency traders increase net gains from trading

activity. These results would indicate that the Bollinger Band strategy would produce larger

negative returns and should not be considered by currency trading managers for implementation.

In conclusion, the null hypothesis was rejected for hypothesis 1 and hypothesis 2, and the

alternative hypothesis was accepted for both hypotheses. However, the results of the Bollinger

89
Band strategy produced a mean net negative gain from trading activity. This would indicate that

using the alternative trading strategy of Bollinger Bands on an intraday currency trading activity

produces a negative mean net gain that is significantly different than the negative mean net gains

of the buy and hold strategy. Furthermore, using the alternative trading strategy of Bollinger

Bands on an intraday currency trading activity produces a negative return that is significantly

different than the positive mean net gains of the sell and hold strategy.

Relationship of the hypotheses to research questions. The research for this study was

designed to address the research question: Do currency trading managers at international banks

and hedge funds experience a difference in net gain from currency transactions when employing

a traditional trading approach as compared to an alternative trading strategy when trading the

EUR/USD spot market on an intraday period? Hypothesis one examined the difference between

the alternative trading strategy using Bollinger Bands and the traditional buy and hold strategy.

Based upon the sample used in this study, it was determined that there is a statistically significant

difference between the two trading strategies. However, the alternative strategy using Bollinger

Bands produced a mean negative net gain. This would indicate that currency traders and

managers of currency traders should not expect the alternative strategy using Bollinger Bands for

intraday currency trading to increase net gains.

Hypothesis two examined the difference between the alternative trading strategy using

Bollinger Bands and the sell and hold strategy. Based upon the sample used in this study, it was

determined that there is a statistically significant difference between the two trading strategies.

However, the alternative strategy using Bollinger Bands produced a mean negative net gain.

This would indicate that currency traders and managers of currency traders should not expect the

alternative strategy using Bollinger Bands for intraday currency trading to increase net gains.

90
Moreover, the sell and hold strategy did produce a mean positive net gain over the sample period

and should be considered by currency traders and currency managers as a method for increasing

net gains from trading activities.

Additional findings. Although not a part of the original research question, after

performing the initial ANOVA test and appropriate t-tests, this researcher wanted to see if there

were statistical differences between the three strategies on a year-by-year basis. An ANOVA

test was performed for each year and the results can be found in Appendix C. The p-value for of

the ANOVA test for 2009 was 0.0245 indicating that there is a statistically significant difference

between the average net gains from the three different trading techniques. Conducting a post-

hoc t-Test to determine where the difference exists, the p-value of 0.013 was obtained for the

difference with the Bollinger Band and buy and hold strategy. A p-value of 0.38 was obtained

for the difference between the mean net gain generated from the Bollinger Band strategy and the

sell and hold strategy. Therefore, the difference with the Bollinger Band and the buy and hold

strategy is the only difference that is statistically significant.

The p-value from the ANOVA test for the years 2010, 2011, 2012, 2013, and 2014 did

not produce any statistically significant difference between the average net gains produced from

the three strategies. The lowest p-value obtained from 2010 through 2014 was 0.3314. The

ANOVA results for all years can be found in Appendix C. The ANOVA produced a p-value of

0.0000 for 2015 and a p-value of 0.0001 for 2016 indicating that there is a statistically significant

difference between the mean net gains produced by each trading strategy.

For 2015, a post-hoc t-Test was performed for the Bollinger Band and buy and hold

strategy. The t-Test produced a p-value of 0.0000000000853 which is less than the alpha of .10

indicating a statistically significant difference between the two strategies. However, the outcome

91
of the Bollinger Band strategy resulted in a negative net gain which does not support the use of

this strategy as a method to increase net gains from intraday trading of the EUR/USD.

Additionally, a post-hoc t-Test was performed for the Bollinger Band and sell and hold strategy.

The t-Test produced a p-value of 0.0000000000886, which is less than the alpha of .10 indicating

a statistically significant difference between the two strategies. However, the outcome of the

Bollinger Band strategy resulted in a negative net gain which does not support the use of this

strategy as a method to increase net gains from intraday trading of the EUR/USD. Finally, a t-

Test was performed using the buy and hold strategy and the sell and hold strategy. The p-value

for this test was 0.9670, which is greater than the level of alpha of .10 indicating there is no

significant difference between the buy and hold strategy and the sell and hold strategy. This

indicates the statistical difference lies between the Bollinger Band strategy and buy and hold

strategy and the Bollinger Band Strategy and the sell and hold strategy. This also indicates there

is no statistically significant difference between the buy and hold strategy and the sell and hold

strategy. The full results of the t-Test for 2015 can be found in Appendix D.

For 2016, a post-hoc t-Test was performed for the Bollinger Band and buy and hold

strategy. The t-Test produced a p-value of .0017 which is less than the alpha of .10 indicating a

statistically significant difference between the two strategies. However, the outcome of the

Bollinger Band strategy resulted in a negative net gain which does not support the use of this

strategy as a method to increase net gains from intraday trading of the EUR/USD. Additionally,

a post-hoc t-Test was performed for the Bollinger Band and sell and hold strategy. The t-Test

produced a p-value of .0029, which is less than the alpha of .10 indicating a statistically

significant difference between the two strategies. Although there was a statistically significant

difference, the outcome of the Bollinger Band strategy resulted in a negative net gain which does

92
not support the use of this strategy as a method to increase net gains from intraday trading of the

EUR/USD. Finally, a t-Test was performed using the buy and hold strategy and the sell and hold

strategy. The p-value for this test was 0.3544, which is greater than the level of alpha of .10

indicating there is no significant difference between the buy and hold strategy and the sell and

hold strategy. The full results of the t-Test for 2016 can be found in Appendix D.

Summary of Findings

Currently, there is gap in the literature relating to intraday trading, tick-level data, and

technical analysis (Ozturk et al., 2016). Much of the current literature focuses on the end of day

price changes of currency pairs and does not focus on intraday price changes (Hayes et al., 2016;

Narayan et al., 2015; Poti et al., 2014). Additionally, most studies that have tested Bollinger

Bands and other alternative trading methods focused on individual stocks, ETFs, or indices

without much focus on specific foreign currency trading (Brock et al., 1992; Lento et al., 2007;

Duvinage et al., 2013; Metghalchi et al., 2014). This study attempted to add to the scholarly

body of knowledge by focusing on intraday price changes and the application of Bollinger Bands

to currency trading.

The overall mean produced by the Bollinger Band strategy, an alternative trading method,

for the sample period was negative. The descriptive statistics on a year-by-year basis showed the

Bollinger Band strategy did produce positive mean net gains in 2009, 2010, and 2014, however

the negative net losses in 2015 and 2016 far exceeded any gains in previous years. A test for

normality was also conducted. The following normality tests were performed for each of the

sample data using Microsoft Excel and an add-on data analysis tool provided by NumXL: the

Jarque-Bera, the Sharipo-Wilk, and the Doornik Chi-Squared. At the alpha of 0.05, all sample

data for each year passed the test of normality.

93
The Bollinger Band trading strategy showed significantly different mean negative net

gains as compared to the buy and hold strategy and the mean positive net gain from the sell and

hold strategy. Therefore, both null hypotheses were rejected at an alpha level of 0.05 and the

alternative hypotheses were accepted. The post-hoc power test using G*Power 3.1 provided a

power level of .9839 for the buy and hold strategy as compared to Bollinger Band strategy and a

power level of .9937 for the sell and hold strategy as compared to the Bollinger Band strategy.

The outcome of the Bollinger Band strategy resulted in a negative net gain which does not

support the use of this strategy as a method to increase net gains from intraday trading of the

EUR/USD. Moreover, the sell and hold strategy did produce a mean positive net gain over the

sample period and should be considered by currency traders and currency managers as a method

for increasing net gains from trading activities.

Applications to Professional Practice

Increased global commerce has given rise to the importance of foreign currency exchange

markets (Harvey, 2013). Multinational firms engage in business activities that are outside of

their home country and often adapt products, services, and pricing to each local market to be

successful (Devereux et al., 2017). The local currency is the easiest way for local consumers to

make purchases from international firms (Satterlee, 2014). These global companies then rely on

the services of international banks to exchange the foreign currency for the currency of the home

country. It is this trend in a global economic expansion that drives currency transactions at

global financial institutions (Cetorelli & Goldberg, 2012; Meschi et al., 2016).

Currency trading is an essential function of all international businesses, a key service for

international banks, and a vital source of profitability for international hedge funds (Wong,

2016). The findings and results produced from this research may be applied by those working in

94
the currency markets. Specifically, the results from this study are applicable for currency traders

working for international financial institutions and currency managers overseeing currency

trading operations.

Currency Manager and Currency Trader

International financial institutions act as brokers for their global clients in the currency

market which causes these global financial institutions, as well as hedge funds, to speculate in

currency markets as a means to increase profits of their existing currency positions (Aktan et al.,

2013; Shen & Hartarska, 2013). Many currency trading teams and currency managers are

seeking for a competitive advantage in the spot currency market that will increase net gains for

their financial firm (Bahmani-Oskooee & Aftab, 2017; Kamau et al., 2015). As cited previously

in the literature review, currency managers and currency traders working for these financial firms

are constantly looking for tools to help increase net gain from currency activity. The findings of

this study indicate that currency traders using Bollinger Bands at a tick-level while trading the

EUR/USD from 2009 to 2016 probably would not have increased the net gains from trading

activity. Based upon the findings of this study, it can be reasoned that using Bollinger Bands at a

tick level within intraday currency trading would not increase net gains from currency trading.

Biblical Implications

The biblical principle of stewardship should be evident in all areas of life, including

international currency transactions. The biblical view of stewardship is traditionally defined as

using and managing the resources God provides for the glory of God and the betterment of his

creation (Botha, 2014). Stewardship can further be understood as entrusting someone with

resources and acting on behalf of another’s interest (McCuddy & Pirie, 2007). All people are

responsible for the resources they control, and wasted resources quickly yield no fruit (McCuddy

95
& Pirie, 2007). Currency traders working for hedge funds and international banks have a

responsibility to make decisions that are in the best interest of the business owners and clients of

the firm (Morck, 2014) or risk producing no fruit for their owners. By utilizing the best tools

available, currency traders can potentially maximize value for clients and owners (Osiyevskyy &

Biloshapka, 2017), which is foundational to stewardship (Luke 16:2, MEV). Understanding the

effectiveness of a trading tool like Bollinger Bands can increase the return and minimize the risk

for currency traders. This leads to increased stewardship by currency managers and currency

traders (Luke 19:12-19, NASB). The findings of this study indicate that using Bollinger Bands

for trading tick-level data does not increase the net gain from trading activity. Currency

managers and traders aware of these findings are able to practice stewardship by ensuring the

greatest net gains for clients and owners by not using Bollinger Bands on tick-level data.

Finally, the role of the global financial institutions should fit into God’s design (Van

Doser, 2010). “And there are varieties of ministries, and the same Lord. There are varieties of

effects, but the same God who works all things in all persons. But to each one is given the

manifestation of the Spirit for the common good” (1 Corinthians 12:5-7, NASB). Global

financial traders and business professionals can fulfill God’s purpose by providing resources for

God’s creation to flourish with the understanding that these goods and services are beneficial to

God’s people (Van Doser, 2010). Positive net gains are vital to any for-profit organization, but

within God’s design, making money provides a means to serve God’s creation rather than being

the ultimate pursuit of the organization (Van Duzer, 2010). All people are responsible for the

resources they control, and wasted resources quickly yield no fruit (McCuddy & Pirie, 2007).

There, implementing effective biblically based stewardship within the foreign exchange markets

96
for international banks and hedge funds includes the exploration of alternative investment

methods.

Recommendations for Action

The results of this research could potentially be of value to managing currency traders

and for currency traders themselves. Additionally, the findings could potentially be of value to

all those who engage in currency trading activity on a speculative basis.

The results of this study suggest that the use of alternative trading strategies, specifically

Bollinger Bands as tested and applied to tick-level time frames did not result in an increase net

gains as compared to traditional buy and hold trading strategies or the sell and hold trading

strategy. Although three of the years included in the sampling showed a positive net gain from

the Bollinger Band strategy, five of the sample years showed negative net gains. Furthermore,

two years out of the five years that showed a negative net gain produced exceedingly large losses

as compared to the more conservative trading strategies. This suggests that the use of Bollinger

Bands may or may not be useful for currency traders and currency trading managers as they did

not provide a positive net gain during the sample period in this study.

Currency traders desiring to use Bollinger Bands in their trading should do so cautiously,

based upon the findings from this study. The findings of this study did not indicate the use of

Bollinger Bands at the tick-level using 3 standard deviations from a 20 period moving average as

signals for entering and exiting currency trades. Although the results presented in this study

show the net gains from the Bollinger Band strategy for 2009 were statistically different than

those of the buy and hold strategy, there was no statistical difference between the sell and hold

strategy and this was the only year out of the eight-year sample which showed a difference. The

information put forth in this research study should be examined by those working as currency

97
traders in international banks and hedge funds. However, the findings should also provide a

catalyst for additional research.

Recommendations for Further Study

This research contributed to the current gap in the literature concerning the use of

Bollinger Bands on an intraday basis, specifically at the tick-level. Additional research in this

area is warranted. The following are recommendations for further research.

First, the sample size for future studies could be increased. A larger sample size would

result in a higher confidence level and a lower confidence interval. The study could be repeated

with similar parameters but increase the sample size to 360 days. Although the power test

showed a significant effect, a larger sample size will add additional confidence. Furthermore,

future studies might include every trading day instead of sampling specific days to better

understand the Bollinger band strategy on the entire population of trading days.

Second, there are other timeframes that should be considered as an area for further study.

Currency traders might focus their attention on timeframes of 1-minute, 5-minute, 10-minute,

15-minute or even 30-minute trading. These various timeframes would then produce different

moving averages from which to calculate the standard deviation prices. These different standard

deviation prices would be used for trade entry signals and trade exit signals. Using longer

timeframes would result in fewer trades and fewer transaction costs as compared to the Bollinger

Band strategy employed in this study.

Third, a further study might examine which Bollinger Band parameters provide the best

results for tick-level data. This study used a 20-period moving average and 3 standard deviations

as the settings to create the Bollinger Bands. Various standard deviations ranging from one to

five might produce different results over the sample period. Additionally, one could argue that a

98
20-period moving average on tick-level data is not a long enough time period to establish a trend.

Extending the number of periods in the moving average to 100 or even 200 ticks might produce

different results than those found in this study.

Fourth, additional studies could focus on different currency pairs and different time

periods. This study focuses exclusively on the EUR/USD, given the significance of these two

currency pairs in the global foreign exchange market as previously cited. However, applying the

Bollinger Bands to other currency pairs over the same time period might produce different

results. Additional factors of liquidity and volume might play a factor in the results.

Lastly, combining the Bollinger Band strategy with another alternative trading strategy

on a tick-level timeframe could also be an area for further study. This study focused on

Bollinger Bands for both entering trades and exiting trades. However, there is a potential to

combine the Bollinger Band strategy with a relative strength indicator or a moving average

convergence divergence strategy for entering and exiting a trade might produce different results

and reduce transaction costs using alternative trading strategies.

Reflections

As an active day trader in the U.S. equity markets and a user of technical analysis, I

began this study with a desire to better understand if an increase in net gains could be realized

using alternative trading strategies in the currency market. The results from this quantitative

study indicated that currency traders using Bollinger Bands on a tick-level should do so

cautiously. Additionally, the results provided opportunities for further studies using Bollinger

Bands with various parameters, different currency pairs, and with larger sample sizes.

If the results of the Bollinger Band strategy matched or outperformed the net gains from

the buy and hold strategy and the sell and hold strategy, the researcher would have considered

99
the study to have a positive outcome for those using technical analysis. However, the findings

presented here do not support the use of Bollinger Bands.

For this study the researcher relied on historical data obtained from a third party. Using

historical data helped to eliminate research bias and add to the validity and reliability of the

results (Rudestam & Newton, 2015). If the Bollinger Bands had produced a statistically

significant difference in positive net gains than the other two traditional strategies or if the mean

gains were equal, then the currency trader could consider using this strategy as part of his or her

daily analysis of the currency market.

Although the findings indicate that Bollinger Bands should be used cautiously for tick-

level trading of the EUR/USD, the opinion of the researcher is this research provided important

insights for currency traders and currency trading managers working for international financial

institutions and global hedge funds. This study helped to provide awareness for the potential

negative net gains for those considering the use of Bollinger Bands on an intraday basis,

specifically using them for tick-level trading. The researcher thinks there are currency traders

that will continue to use Bollinger Bands in their daily currency trading, but will most likely

adjust the parameters of the Bollinger Bands to adapt in real time to the changing market

conditions.

Summary and Study Conclusions

For this study the researcher examined the net gain and losses of currency transactions

using alternative trading techniques that are available for currency traders and currency managers

working at international banks and hedge funds. The specific business problem addressed by

this study was to provide currency trading manager’s additional data to help determine the most

effective tool to increase net gains from trading activity. The results of this study indicate that

100
using Bollinger Bands while trading the EUR/USD spot market at the tick-level probably does

not increase the net gains from trading activity.

The Bollinger Band strategy, an alternative trading method, produced an overall mean

negative net gain for the sample period. The descriptive statistics on a year-by-year basis

showed the Bollinger Band strategy did produce positive mean net gains in 2009, 2010, and

2014; however, the negative net losses in 2015 and 2016 far exceeded any gains in previous

years. A test for normality was also conducted. The following normality tests were performed

for each of the sample data using Microsoft Excel and an add-on data analysis tool provided by

NumXL: the Jarque-Bera, the Sharipo-Wilk, and the Doornik Chi-Squared. At the alpha of 0.05,

all sample data for each year passed the test of normality.

The Bollinger Band trading strategy showed significantly different mean negative net

gains as compared to the buy and hold strategy and the mean positive net gain from the sell and

hold strategy. Therefore, both null hypotheses were rejected at an alpha level of 0.10 and the

alternative hypotheses were accepted. This would indicate that currency traders and managers of

currency traders should not expect the alternative strategy using Bollinger Bands for intraday

currency trading to increase net gains. Moreover, the sell and hold strategy did produce a mean

positive net gain over the sample period and should be considered by currency traders and

currency managers as a method for increasing net gains from trading activities.

These findings have a direct application to the business problem of increasing the net

gain from foreign currency transactions undertaking by currency management teams and

currency traders within global financial institutions and hedge funds. Additionally, the results of

this study have added to the current literature on technical analysis and have helped to fill in the

gap with regards to tick-level time frames.

101
References

Aabo, T., Pantzalis, C., & Park, J. C. (2017). Idiosyncratic volatility: An indicator of noise

trading? Journal of Banking and Finance, 75, 136-151. doi:

10.1016/j.jbankfin.2016.11.003

Abby, B. S., & Doukas, J. A. (2012). Is technical analysis profitable for individual currency

traders? Journal of Portfolio Management, 39(1), 142.

Abbey, B. S., & Doukas, J. A. (2015). Do individual currency traders make money? Journal of

International Money and Finance, 56, 158-177. doi: 10.1016/j.jimonfin.2014.10.003

Abildgren, K. (2014). Tail events in the FX markets since 1740. The Journal of Risk Finance,

15(3), 294-311. doi: 10.1108/JRF-04-2014-0041

Abuhamad, M., Mohd, M., & Salim, J. (2013). Event driven business intelligence approach for

real-time integration of technical and fundamental analysis in forex market. Journal of

Computer Science, 9(4), 488-499.

Ackerman, D. T. (2016). What is forex and is more regulation necessary? This is the short and

the long of it. Banking & Financial Services Policy Report, 35(9), 11-21.

Addam, O., Chen, A., Hoang, W., Rokne, J., & Alhajj, R. (2016). Foreign exchange data

crawling and analysis for knowledge discovery leading to informative decision making.

Knowledge-Based Systems, 102, 1-19. doi: 10.1016/j.knosys.2016.03.005

Ahmad, M. I., Guohui, W., Rafiq, M. Y., Hasan, M., Chohan, A., & Sattar, A. (2017). Assessing

performance of moving average investment timing strategy over the UK stock market.

Journal of Developing Areas, 51(3), 349.

Aktan, B., Chan, S. G., Žiković, S., & Evrim-Mandaci, P. (2013). Off-balance sheet activities

impact on commercial banks performance: An emerging market perspective. Economic

102
Research-Ekonomska Istraživanja, 26(3), 117-132. doi:

10.1080/1331677X.2013.11517625

Allayannis, G., Ihrig, J. & Weston, J. (2001). Exchange-rate hedging: financial vs operational

strategies. American Economic Review Papers and Proceedings, 91(2), 391-395.

Alvarez-Ramirez, J., Rodriguez, E., & Espinosa-Paredes, G. (2012). Is the US stock market

becoming weakly efficient over time? Evidence from 80-year-long data. Physica A:

Statistical Mechanics and its Applications, 391(22), 5643. doi:

10.1016/j.physa.2012.06.051

Andersen, T. G., & Bollerslev, T. (1997). Intraday periodicity and volatility persistence in

financial markets. Journal of Empirical Finance, 4(2), 115-158. doi: 10.1016/S0927-

5398(97)00004-2

Andersen, T. G., Bollerslev, T., Diebold, F. X., & Vega, C. (2003). Micro effects of macro

announcements: Real time price discovery in foreign exchange. American Economic

Review, 93(1), 38-62.

Andersen, T. G., Bollerslev, T., Diebold, F. X., & Vega, C. (2007). Real-time price discovery in

global stock, bond and foreign exchange markets. Journal of International Economics,

73(2), 251-277. doi: 10.1016/j.jinteco.2007.02.004

Anderson, J., & Faff, R. (2005). Profitability of trading rules in futures markets. Accounting

Research Journal, 18(2), 83-92. doi: 10.1108/10309610580000677

Antinolfi, G., & Kawamura, E. (2008). Banks and markets in a monetary economy. Journal of

Monetary Economics, 55 (2), 321-334. doi: 10.1016/j.jmoneco.2007.12.009

Anzuini, A., & Fornari, F. (2012). Macroeconomic determinants of carry trade activity. Review

of International Economics, 20, 468-488.

103
Arnoldi, J. (2016). Computer algorithms, market manipulation and the institutionalization of

high frequency trading. Theory, Culture & Society, 33(1), 29-52. doi:

10.1177/0263276414566642

Azizan, N. A., & M’ng, J. C. P. (2010). Can technical analysis predict the movement of futures

prices? Journal of Financial Risk Management, 7(3), 57.

Bacchetta, P., & van Wincoop, E. (2013). On the unstable relationship between exchange rates

and macroeconomic fundamentals. Journal of International Economics, 91(1), 18-26.

Bachelier, L. (1900). Theorie de la speculation. Gauthiers-Villars, Paris. Reprinted in English

(translated by Boness, A.J.). In P. H. Coother (Ed.), The random character of stock

market prices (pp. 17-87). Cambridge, MA: MIT Press, 1964.

Bahmani-Oskooee, M., & Aftab, M. (2017). On the asymmetric effects of exchange rate

volatility on trade flows: New evidence from US-Malaysia trade at the industry level.

Economic Modelling, 63, 86-103. doi: 10.1016/j.econmod.2017.02.004

Baillie, R. T., & Chang, S. S. (2011). Carry trades, momentum trading and the forward premium

anomaly. Journal of Financial Markets, 14, 441-464.

Bajgrowicz, P., & Scaillet, O. (2012). Technical trading revisited: false discoveries, persistence

tests, and transaction costs. Journal of Financial Economics, 106(3), 473-491.

Bakshi, G., & Panayotov, G. (2013). Predictability of currency carry trades and asset pricing

implications. Journal of Financial Economics, 110, 139-163.

Bank of International Settlement. (2016, April). Turnover of OTC foreign exchange instruments.

Retrieved from https://www.bis.org/statistics/d11_1.pdf.

Baranga, L. P. (2016). Study on the new financial products traded on the forex market. Bulletin

of the Transilvania University of Brasov. Economic Sciences. Series V, 9(1), 179-188.

104
Barunik, J., Krehlik, T., & Vacha, L. (2016). Modeling and forecasting exchange rate volatility

in time-frequency domain. European Journal of Operational Research, 251(1), 329-340.

doi: 10.1016/j.ejor.2015.12.010

Bastidon, C. (2017). Stock markets fragmentation, volatility and final investors. Annals of

Finance, 13(4), 435-451. doi: 10.1007/s10436-017-0305-0

Beilis, A., Dash, J. W., & Wise, J. V. (2014). Psychology, Stock/FX trading and option prices.

Journal of Behavioral Finance, 15(3), 251-268. doi: 10.1080/15427560.2014.943227

Bekiros, S. D. (2014). Exchange rates and fundamentals: Co-movement, long-run relationships

and short-run dynamics. Journal of Banking and Finance, 39, 117-134. doi:

10.1016/j.jbankfin.2013.11.007

Belaire-Franch, J., & Opong, K. K. (2005). Some evidence of random walk behavior of Euro

exchange rates using ranks and signs. Journal of Banking and Finance, 29, 1631-1643.

Ben Omrane, W., & Hussain, S. M. (2016). Foreign news and the structure of co-movement in

European equity markets: An intraday analysis. Research in International Business and

Finance, 37, 572-582. doi: 10.1016/j.ribaf.2016.01.021

Bitvai, Z., & Cohn, T. (2015). Day trading profit maximization with multi-tasking learning and

technical analysis. Machine Learning, 101, 187-209. doi: 10.1007/s10994-014-5480-x

Bleaney, M., & Li, Z. (2016). Decomposing the bid–ask spread in multi‐dealer markets.

International Journal of Finance & Economics, 21(1), 75-89. doi: 10.1002/ijfe.1533

Blume, L., Easley, D., & O’Hara, M. (1994). Market statistics and technical analysis: The role of

volume. Journal of Finance, 49(1), 153-181.

Bollinger, J. (2002). Bollinger on Bollinger Bands. New York, NY: McGraw-Hill.

105
Bonenkamp, U., Homburg, V., & Kempf, A. (2011). Fundamental information in technical

trading strategies. Journal of Business Finance & Accounting, 38(7-8), 842-860.

Boudoukh, J., Richardson, M., & Whitelaw, R. (2008). The myth of long-horizon predictability.

The Review of Financial Studies, 21(4), 1577-1605.

Brauer, G. A. (1988). Closed-end fund shares' abnormal returns and the information content of

discounts and premiums. The Journal of Finance, 43(1), 113-127. doi: 10.1111/j.1540-

6261.1988.tb02592.x

Breedon, F., & Ranaldo, A. (2013). Intraday patterns in FX returns and order flow. Journal of

Money, Credit and Banking, 45(5), 953-965. doi: 10.1111/jmcb.12032

Bréhier, B. (2013). High frequency trading: Should technological developments be considered a

potential threat to financial markets and be subject to specific regulation? ERA Forum,

14(1), 69-80. doi: 10.1007/s12027-013-0293-0

Brock, A. W., Hommes, C. H., & Wagener, F.O.O. (2009). More hedging instruments may

destabilize markets. Journal of Economic Dynamic Control, 33, 1912-1928.

Brock, W., Lakonishok, J., & LeBaron, B. (1992). Simple technical trading rules and the

stochastic properties of stock returns. Journal of Finance, 47(5), 1731-1764.

Burnside, C., Eichenbaum, M., & Rebelo, S. (2008). Carry trade: The gains of diversification.

Journal of the European Economic Association, 6, 581-588.

Bush, P. J., & Stephens, J. E. (2016). The Monday effect in the EUR/USD currency pair: Periods

of EUR strength and weakness. Journal of Behavioral and Experimental Finance, 10, 72-

74. doi: 10.1016/j.jbef.2016.03.003

106
Caporale, G. M., Gil-Alana, L., & Plastun, A. (2018). Short-term price overreactions:

Identification, testing, exploitation. Computational Economics, 51(4), 913-940. doi:

10.1007/s10614-017-9651-2

Cash, P., Stankovic, T., & Storga, M. (2016). Experimental design research. Switzerland:

Springer International Publishing.

Cenedese, G., Sarno, L., & Tsiakas, I. (2014). Foreign exchange risk and the predictability of

carry trade returns. Journal of Banking and Finance, 42, 302-313.

Cerreia-Vioglio, S., Maccheroni, F., & Marinacci, M. (2015). Put-call parity and market

frictions. Journal of Economic Theory, 157, 730-762. doi: 10.1016/j.jet.2014.12.011

Cetorelli, N., & Goldberg, L. S. (2012). Banking globalization and monetary transmission. The

Journal of Finance, 67(5), 1811-1843.

Chadwick, M. G., Fazilet, F., & Tekatli, N. (2015). Understanding the common dynamics of the

emerging market currencies. Economic Modelling, 49, 120-136. doi:

10.1016/j.econmod.2015.03.011

Chang, C., Huang, H., Chang, Y., & Lin, T. (2015). Stock characteristics, trading behavior, and

psychological pitfalls. Managerial Finance, 41(12), 1298-1317. doi: 10.1108/MF-02-

2014-0053

Chang, K. P. H., & Osler, C. L. (1999). Methodical madness: Technical analysis and the

irrationality of exchange- rate forecasts. The Economic Journal, 109(458), 636-661. doi:

10.1111/1468-0297.00466

Chang, Y., Jong, C., & Wang, S. (2017). Size, trading volume, and the profitability of technical

trading. International Journal of Managerial Finance, 13(4), 475.

107
Charles, A., Darne, O., & Kim, J. H. (2012). Exchange-rate predictability and adaptive market

hypothesis: Evidence from major exchange rates. Journal of International Money and

Finance, 31(6), 1607-1626. doi: 10.1016/j.jimonfin.2012.03.003.

Chen, J. H., & Diaz, J. F. (2013). Long memory and shifts in the returns of green and non-green

exchange-traded funds. International Journal of Humanities and Social Science

Invention, 2(10), 29-32.

Chen, S., Bao, S., & Zhou, Y. (2016). The predictive power of Japanese candlestick charting in

Chinese stock market. Physica A: Statistical Mechanics and its Applications, 457, 148-

165. doi: 10.1016/j.physa.2016.03.081

Chen, S., Chen, N., & Chuang, R. (2014). An empirical study on technical analysis: GARCH (1,

1) model. Journal of Applied Statistics, 41(4), 785-801. doi:

10.1080/02664763.2013.856383

Chen, W. D., & Li, H. C. (2016). Wavelet decomposition of heterogeneous investment horizon.

Journal of Economics and Finance, 40(4), 714-734. doi: 10.1007/s12197-015-9321-y

Chiarella, C., & Ladley, D. (2016). Chasing trends at the micro-level: The effect of technical

trading on order book dynamics. Journal of Banking & Finance, 72, S119-S131.

Chien, C., Lee, H., Tai, S., & Liao, T. (2013). Information, hedging demand, and institutional

investors: Evidence from the Taiwan futures exchange. Journal of Multinational

Financial Management, 23(5), 394-414. doi: 10.1016/j.mulfin.2013.08.001

Chong, L., Chang, X., & Tan, S. (2014). Determinants of corporate foreign exchange risk

hedging. Managerial Finance, 40(2), 176-188. doi: 10.1108/MF-02-2013-0041

Clare, A., Seaton, J., Smith, P. N., & Thomas, S. (2014). Trend following, risk parity and

momentum in commodity futures. International Review of Financial Analysis, 31, 1-12.

108
Coakley, J., Marzano, M., & Nankervis, J. (2016). How profitable are FX technical trading

rules? International Review of Financial Analysis, 45, 273-282. doi:

10.1016/j.irfa.2016.03.010

Coe, T. S., & Laosethakul, K. (2010), Should individual investors use technical trading rules to

attempt to beat the market? American Journal of Economic and Business Administration,

2(3), 201-209.

Cohen, G., & Cabiri, E. (2015). Can technical oscillators outperform the buy and hold strategy?

Applied Economics, 47(30), 3189-3197. doi: 10.1080/00036846.2015.1013609

Cohen, J. (1998). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ:

Lawrence Earlbaum Associates.

Colavecchio, R. (2008). Tracking the yen carry trade: Evidence from a regime switching

approach. Available at SSRN: http://ssrn.com/abstract=1365968.

Condie, S., & Ganguli, J. (2011). Ambiguity and rational expectations equilibria. Review of

Economic Studies, 78(3), 821-845.

Copeland, L., & Lu, W. (2016). Dodging the steamroller: Fundamentals versus the carry trade.

Journal of International Financial Markets, Institutions & Money, 42, 115-131. doi:

10.1016/j.intfin.2016.02.004

Cornell, W. B., & Dietrich, J. K. (1978). The efficiency of the market for foreign exchange under

floating exchange rates. Review of Economic Statistics, 60, 111-120.

Costantini, M., Cuaresma, J. C., & Hlouskova, J. (2016). Forecasting errors, directional accuracy

and profitability of currency trading: The case of EUR/USD exchange rate. Journal of

Forecasting, 35(7), 652-668. doi: 10.1002/for.2398

109
Coudert, V., & Mignon, V. (2013). The “forward premium puzzle” and the sovereign default

risk. Journal of International Money and Finance, 32, 491-511.

Creswell, J. W. (2014). Research design, qualitative, quantitative, and mixed methods

approaches (4th ed.). Los Angeles, CA: Sage.

Crowder, W. J. (2014). Real exchange rate persistence in US dollar PPP systems. International

Finance, 17(2), 209-240. doi: 10.1111/infi.12048

Curato, G., & Lillo, F. (2014). Multiscale model selection for high-frequency financial data of a

large tick stock by means of the Jensen–Shannon metric. Entropy, 16(1), 567-581. doi:

10.3390/e16010567.

da Costa, T. R. C. C., Nazário, R. T., Bergo, G. S. Z., Sobreiro, V. A., & Kimura, H. (2015).

Trading system based on the use of technical analysis: A computational experiment.

Journal of Behavioral and Experimental Finance, 6, 42-55.

Dal Bianco, M., Camacho, M., & Perez Quiros, G. (2012). Short-run forecasting of the euro-

dollar exchange rate with economic fundamentals. Journal of International Money and

Finance, 31(2), 377-396. doi: 10.1016/j.jimonfin.2011.11.018

Daniel, K., & Titman, S. (1999). Market efficiency in an irrational world. Financial Analysts

Journal, 55(6), 28-40.

Das, S., Kadapakkam, P. R., & Tse, Y. (2013). Is carry-trade a viable alternative asset class?

Journal of International Financial Markets, Institutions and Money, 24, 247-257.

Davis, P. (2017). Value investing: Do quant strategies measure up? Financial Analysts Journal,

73(2), 53.

De Bondt, W. F., & Thaler, R. (1985). Does the Stock Market Overreact? The Journal of

Finance, 40(3), 793-805.

110
Degutis, A., & Novickyte, L. (2014). The efficient market hypothesis: A critical review of

literature and methodology. Ekonomika, 93(2), 7.

Deng, S., & Sakurai, A. (2013). Foreign exchange trading rules using single technical indicator

from multiple timeframes. Advanced Information Networking, 3, 207-212.

Detollenaere, B., & Mazza, P. (2014). Do Japanese candlesticks help solve the trader’s dilemma?

Journal of Banking & Finance, 48, 386-395.

Devereux, M. B., Dong, W., & Tomlin, B. (2017). Importers and exporters in exchange rate

pass-through and currency invoicing. Journal of International Economics, 105, 187-204.

doi: 10.1016/j.jinteco.2016.12.011

de Zwart, G., Markwat, T., Swinkels, L., & van Dijk, D. (2009). The economic value of

fundamental and technical information in emerging currency markets. Journal of

International Money and Finance, 28(4), 581-604. doi: 10.1016/j.jimonfin

.2009.01.004

Dhankar, R. S., & Shankar, D. (2016). Relevance and evolution of adaptive markets hypothesis:

A review. Journal of Indian Business Research, 8(3), 166-179.

Dobrynskaya, V. (2015). Currency exposure to downside risk: Which fundamentals matter?:

Currency exposure to downside risk. Review of International Economics, 23(2), 345-360.

doi: 10.1111/roie.12174

Dooley, M. P., & Shafer, J. (1975). Analysis of short run exchange rate behavior. International

Finance Discussion Papers, 76, 34-49.

Dowell-Jones, M. (2012). Investing in emerging markets: The ethical context. The Journal of

Corporate Citizenship, (48), 75-90.

111
Drachal, K. (2016). Intraday volatility and var: An evidence from the construction sector.

Urbanism. Arhitectura. Constructii, 7(4), 349-358.

Drechsler, I. (2013). Uncertainty, time-varying fear, and asset prices. The Journal of Finance,

68(5), 1843-1889. doi: 10.1111/jofi.12068

Duffie, D., Gârleanu, N., & Pedersen, L. H. (2007). Valuation in over-the-counter markets. The

Review of Financial Studies, 20(6), 1865-1900. doi: 10.1093/rfs/hhm037

Dunis, C. L., Laws, J., & Evans, B. (2006). Trading futures spreads: An application of

correlation and threshold filters. Applied Financial Economics, 16(12), 903-914.

Duvinage, M., Mazza, P., & Petitjean, M. (2013). The intra-day performance of market timing

strategies and trading systems based on Japanese candlesticks. Quantitative Finance,

13(7), 1059-1070.

Edoho, F. M. (2013). Information and communications technologies in the age of

globalization: Challenges and opportunities for Africa. African Journal of Economic

and Management Studies, 4(1), 9-33.

Eiamkanitchat, N., Moontuy, T., & Ramingwong, S. (2017). Fundamental analysis and technical

analysis integrated system for stock filtration. Cluster Computing, 20(1), 883-894. doi:

10.1007/s10586-016-0694-2

Elaut, G., Frömmel, M., & Lampaert, K. (2016). Intraday momentum in FX markets:

Disentangling informed trading from liquidity provision. Journal of Financial Markets.

doi: 10.1016/j.finmar.2016.09.002

Ellis, C. A., & Parbery, S. A. (2005). Is smarter better? A comparison of adaptive, and simple

moving average trading strategies. Research in International Business and Finance,

19(3), 399-411. doi: 10.1016/j.ribaf.2004.12.009

112
El Ouadghiri, I., & Uctum, R. (2016). Jumps in equilibrium prices and asymmetric news in

foreign exchange markets. Economic Modelling, 54, 218-234. doi:

10.1016/j.econmod.2015.12.025

Evans, C., Pappas, P., & Xhafa, F. (2013). Utilizing artificial neural networks and genetic

algorithms to build an algo-trading model for intra-day foreign exchange speculation.

Mathematical and Computer Modelling, 58(5-6), 1249-1266, doi:

10.1016/j.mcm.2013.02.002

Fakhry, B. (2016). A literature review of the efficient market hypothesis. Turkish Economic

Review, 3(3), 431. doi: 10.1453/ter.v3i3.928

Fama, E. F. (1963). Mandelbrot and the stable Paretian hypothesis: I. Introduction. The Journal

of Business (Pre-1986), 36(4), 420.

Fama, E. F. (1965). The behavior of stock-market prices: I. introduction. The Journal of Business

(Pre-1986), 38(1), 34.

Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work.The

Journal of Finance, 25(2), 383-417. doi: 10.1111/j.1540-6261.1970.tb00518.x

Fama, E. F. (1995). Random walks in stock market prices. Financial Analysts Journal, 51(1), 75.

Faul, F. E., Erdfelder, A. G., & Lang, A. (2007). G* power 3: A flexible statistical power

analysis program for the social, behavioral, and biomedical sciences. Behavior Research

Methods, 39(2), 175-191.

Faust, J., Rogers, J. H., Wang, S. Y. B., Wright, J. H. (2007). The high-frequency response of

exchange rates and interest rates to macroeconomic announcements. Journal of Monetary

Economics, 54, 1051-1068.

113
Ferguson, C. J. (2009). An effect size primer: A guide for clinicians and researchers.

Professional Psychology: Research and Practice, 40, 532-538. doi: 10.1037/a0015808

10.1037/a0015808

Fiebig, H. R., & Musgrove, D. P. (2015). Testing for detailed balance in a financial market.

Physica A: Statistical Mechanics and its Applications, 427, 26-33. doi:

10.1016/j.physa.2015.01.055

Fiedor, P., & Holda, A. (2016). Information-theoretic approach to quantifying currency risk. The

Journal of Risk Finance, 17(1), 93-109.

Fifield, S. G. M., Power, D. M., & Knipe, D. G. S. (2008). The performance of moving average

rules in emerging stock markets. Applied Financial Economics, 18(19), 1515-1532. doi:

10.1080/09603100701720302

Fink, A. (2014). Conducting research literature reviews (4th ed.). Thousand Oaks, CA: Sage.

Fleming, M. J., & Remolona, E. M. (1999). Price Formation and Liquidity in the U.S. Treasury

Market: The Response to Public Information. The Journal of Finance, 54(5),

1901-1915. doi. 10.1111/0022-1082.00172

Flood, R., & Rose, A. (1995). Fixing exchange rates: A virtual quest for fundamentals.

Journal of Monetary Economics, 36, 3-37.

Foltice, B., & Langer, T. (2015). Profitable momentum trading strategies for individual

investors. Financial Markets and Portfolio Management, 29(2), 85-113. doi:

10.1007/s11408-015-0246-4

Fong, W. M. (2013). Footprints in the market: Hedge funds and the carry trade. Journal of

International Money and Finance, 33, 41-59. doi: 10.1016/j.jimonfin.2012.09.001

114
Frömmel, M., & Lampaert, K. (2016). Does frequency matter for intraday technical trading?.

Finance Research Letters, 18, 177-183.

Fu, Y., & Blazenko, G. W. (2015). Returns for dividend paying and non-dividend paying firms

International Journal of Business and Financial Research, 9(2). 1-15.

Fu, Y., & Blazenko, G. W. (2017). Normative portfolio theory. International Review of

Financial Analysis, 52, 240-251. doi: 10.1016/j.irfa.2017.07.002

Fuertes, A., Kalotychou, E., & Todorovic, N. (2015). Daily volume, intraday and overnight

returns for volatility prediction: Profitability or accuracy? Review of Quantitative Finance

and Accounting, 45(2), 251-278. doi: 10.1007/s11156-014-0436-6

Fung, H. G., Tse, Y., & Zhao, L. (2013). Are stock markets in Asia related to carry trade? Pacific

Basin Finance Journal, 25, 200-216.

Galai, D. (1978). Empirical tests of boundary conditions for CBOE options. Journal of Financial

Economics, 6(2), 187-211. doi: 10.1016/0304-405X(78)90029-6

Galariotis, E. C. (2014). Contrarian and momentum trading: A review of the literature. Review of

Behavioral Finance, 6(1), 63-82.

Gencay, R., Ballocchi, G., & Dacorogna, M., Olsen, R., & Pictet, O. (2002). Real-time models

and the statistical properties of foreign exchange rates. International Economic Review,

43, 463-487.

Georgiadis, G., & Mehl, A. (2016). Financial globalization and monetary policy effectiveness.

Journal of International Economics, 103, 200-212. doi: 10.1016/j.jinteco.2016.10.002

Gerber, J. (2016). Currency power: Understanding monetary rivalry. Middletown: American

Library Association dba CHOICE.

115
Gerlein, E. A., McGinnity, M., Belatreche, A., & Coleman, S. (2016). Evaluating machine

learning classification for financial trading: An empirical approach. Expert Systems with

Applications, 54, 193-207. doi: 10.1016/j.eswa.2016.01.018

Ghazani, M. M., & Araghi, M. K. (2014). Evaluation of the adaptive market hypothesis as an

evolutionary perspective of market efficiency: Evidence from the Tehran stock exchange.

Research in International Business and Finance, 32, 50-59.

Ghyselse, E., Hill, J., & Motegi, K. (2013). Testing for granger causality with mixed frequency

data. Journal of Economic Finance, 14, 304-324.

Girma, P. B., & Paulson, A. S. (1998). Seasonality in petroleum futures spreads. Journal of

Futures Market, 18, 581-598.

Girma, P. B., & Paulson, A. S. (1999). Risk arbitrage opportunities in petroleum futures spreads.

Journal of Futures Market, 18, 581-598.

Glabadanidis, P. (2014). The market timing power of moving averages: Evidence from US

REITs and REIT indexes. International Review of Finance, 14(2), 161-202. doi:

10.1111/irfi.12018

Glabadanidis, P. (2015). Market timing with moving averages. International Review of Finance,

15(3), 387-425.

Glabadanidis, P. (2017). Timing the market with a combination of moving averages.

International Review of Finance, 17(3), 353-394. doi: 10.1111/irfi.12107

Gloede, O., & Menkhoff, L. (2014). Financial professionals' overconfidence: Is it experience,

function, or attitude? European Financial Management, 20(2), 236-269. doi:

10.1111/j.1468-036X.2011.00636.x

116
Goldbaum, D. (2003). Profitable technical trading rules as a source of price instability.

Quantitative Finance, 3(3), 220-229.

Gradojevic, N., & Gencay, R. (2013). Fuzzy logic, trading uncertainty and technical trading.

Journal of Banking and Finance, 37(2), 578-586.

Gradojevic, N., & Lento, C. (2015). Multiscale analysis of foreign exchange order flows and

technical trading profitability. Economic Modelling, 47, 156-165. doi:

10.1016/j.econmod.2015.02.028

Grossman, S., & Stiglitz, J. (1976). Information and competitive price systems. American

Economic Review, 66, 246-253.

Grossman, S., & Stiglitz, J. (1980). On the impossibility of informationally efficient markets.

American Economic Review, 70, 393-408.

Hamzaoui, N., & Regaieg, B. (2016). Exploration of the foreign exchange forward premiums

and the spot exchange return: A multivariate approach. International Journal of

Economics and Financial Issues, 6(2), 694-702.

Han, Y., Hu, T., & Yang, J. (2016). Are there exploitable trends in commodity futures prices?

Journal of Banking and Finance, 70, 214-234. doi: 10.1016/j.jbankfin.2016.04.013

Harvey, R. (2013). The legal construction of the global foreign exchange market. Journal of

Comparative Economics, 41(2), 343-354. doi: 10.1016/j.jce.2013.03.006

Hau, H., Massa, M., & Perress, J. (2010). Do demand curves for currencies slope down?

Evidence from the MSCI global index change. Review of Financial Studies, 23, 1681-

1717.

117
Hayes, R. L., Wu, J., Chaysiri, R., Bae, J., Beling, P. A., & Scherer, W. T. (2016). Effects of

time horizon and asset condition on the profitability of technical trading rules. Journal of

Economics and Finance, 40(1), 41-59. doi: 10.1007/s12197-014-9291-5

He, X., & Li, K. (2015). Profitability of time series momentum. Journal of Banking and Finance,

53, 140-157. doi: 10.1016/j.jbankfin.2014.12.017.

Hiremath, G. S., & Kumari, J. (2014). Stock returns predictability and the adaptive market

hypothesis in emerging markets: Evidence from India. SpringerPlus, 3, 428. doi:

10.1186/2193-1801-3-428

Hong, H., & Stein, J. C. (1999). A unified theory of underreaction, momentum trading, and

overreaction in asset markets. The Journal of Finance, 54(6), 2143-2184.

Hong, K. J., & Satchell, S. (2015). Time series momentum trading strategy and autocorrelation

amplification. Quantitative Finance, 15(9), 1471-1487. doi:

10.1080/14697688.2014.1000951

Hou, K., Xue, C., & Zhang, L. (2014). Digesting anomalies: An investing approach. Review of

Financial Studies, 28(3), 650-705.

Hsu, P. H., Taylor, M. P., & Wang, Z. (2016). Technical trading: Is it still beating the foreign

exchange market? Journal of International Economics, 102, 188-208.

Huang, C., Hsu, C., Chen, C., Chang, B. R., & Li, C. (2015). An intelligent model for pairs

trading using genetic algorithms. Computational Intelligence and Neuroscience, 9(3),

960-1006.

Hudson, R., Dempsey, M., & Keasey, K. (1996). A note on the weak form efficiency of capital

markets: The application of simple technical trading rules to UK stock prices 1935-1994.

Journal of Bank Finance, 20, 1121-1132.

118
Hull, M., & McGroarty, F. (2014). Do emerging markets become more efficient as they develop?

Long memory persistence in equity indices. Emerging Markets Review, 18, 45-61.

Hutchison, M., & Sushko, V. (2013). Impact of macro-economic surprises on carry trade

activity. Journal of Banking and Finance, 37, 1133-1147.

Ishida, I., McAleer, M., & Oya, K. (2011). Estimating the leverage parameter of continuous-time

stochastic volatility models using high frequency S&P 500 and VIX. Managerial

Finance, 37(11), 1048-1067. doi: 10.1108/03074351111167938

Ito, M., Noda, A., & Wada, T. (2016). The evolution of stock market efficiency in the US: A

non-bayesian time-varying model approach. Applied Economics, 48(7), 621-635. doi:

10.1080/00036846.2015.1083532

Ito, T., & Hashimoto, Y. (2006). Intraday seasonality in activities of the foreign exchange

markets: Evidence from the electronic broking system. Journal of the Japanese and

International Economies, 20, 637-664. doi: 10.1016/j. jjie.2006.06.005

Jacobs, H., & Hillert, A. (2016). Alphabetic bias, investor recognition, and trading behavior.

Review of Finance, 20(2), 693-723. doi: 10.1093/rof/rfv060

Jawadi, F., Jawadi, N., & Cheffou, A. I. (2015). Are Islamic stock markets efficient? A time-

series analysis. Applied Economics, 47(16), 1686-1697. doi:

10.1080/00036846.2014.1000535

Jensen, M. C. (1978). Some anomalous evidence regarding market efficiency. Journal of

Financial Economics, 6(2-3), 95-101.

Jubinski, D., & Lipton, A. F. (2012). Equity volatility, bond yields, and yield spreads. Journal of

Futures Markets, 32(5), 480-503. doi: 10.1002/fut.20521

119
Jurek, J. W. (2014). Crash-neutral currency carry trades. Journal of Financial Economics, 113,

325-347.

Kablan, A., & Ng, W. L. (2011). Intraday high-frequency FX trading with adaptive neuro-fuzzy

inference systems. International Journal of Finance and Markets Derivatives, 1(2), 68-

87.

Kamau, P., Inanga, E. L, & Rwegasira, K. (2015). The usage of currency derivatives in

multilateral banks. Management Research Review, 38 (5), 482-504. doi: 10.1108/MRR-

10-2013-0248

Kampouridis, M., & Otero, F. E. B. (2017). Heuristic procedures for improving the predictability

of a genetic programming financial forecasting algorithm. Soft Computing, 21(2), 295-

310. doi: 10.1007/s00500-015-1614-8

Kathy, L. (2015). Technical strategy: Trading with double Bollinger Bands. Day Trading and

Swing Trading the Currency Market, 3, 201-215.

Kay, J. (2013). Don’t look for fair prices on the market. Financial Times.

Ke, M., Chiang, Y., & Liao, T. L. (2007). Day-of-the-week effect in the Taiwan foreign

exchange market. Journal of Banking Finance, 31(9), 2847-2865. doi:

10.1016/j.jbankfin.2007.03.005

Khalilzadeh, J., & Tasci, A. D. A. (2017). Large sample size, significance level, and the effect

size: Solutions to perils of using big data for academic research. Tourism Management,

62, 89-96. doi: 10.1016/j.tourman.2017.03.026

Kim, D., & Song, C. (2014). Country fundamentals and currency excess returns. Journal of East

Asian Economic Integration, 18(2), 111-142.

120
Kim, G., An, L., & Kim, Y. (2015). Exchange rate, capital flow and output: Developed versus

developing economies. Atlantic Economic Journal, 43(2), 195-207. doi: 10.1007/s11293-

015-9458-2

Kim, J. H., Shamsuddin, A., & Lim, K. P. (2011). Stock returns predictability and the adaptive

markets hypothesis. Evidence from century long U.S. data. Journal of Empirical Finance,

18, 868-879. doi: 10.1016/j.jempfin.2011.08.002.

Kim, S. (2016). Currency carry trades: The role of macroeconomic news and futures market

speculation. Journal of Futures Markets, 36(11), 1076-1107. doi: 10.1002/fut.21778

Kirilenko, A., Kyle, A. S., Samadi, M., & Tuzun, T. (2017). The flash crash: High‐frequency

trading in an electronic market. The Journal of Finance, 72(3), 967-998. doi:

10.1111/jofi.12498

Kleinnijenhuis, J., Schultz, F., Oegema, D., & Atteveldt, W. (2013). Financial news and market

panics in the age of high-frequency sentiment trading algorithms. Journalism, 14(2), 271-

291.

Kofarbai, H., & Zubairu, M. (2016). Efficient market hypothesis in emerging market - a

conceptual analysis. European Scientific Journal, 12(25). doi:

10.19044/esj.2016.v12n25p260

Kolkova, A. (2017). Testing EMA indicator for the currency pair EUR/USD. International

Journal of Entrepreneurial Knowledge, 5(1), 35-40.

Kolková, A., & Lenertová, L. (2016). Binary options as a modern phenomenon of financial

business. International Journal of Entrepreneurial Knowledge, 4(1), 52-59. doi:

10.1515/ijek-2016-0004

121
Kovalenkov, A., & Vives, X. (2014). Competitive rational expectations equilibria without

apology. Journal of Economic Theory, 149, 211-235.

Krapl, A. A. (2017). Asymmetric foreign exchange cash flow exposure: A firm-level analysis.

Journal of Corporate Finance, 44, 48-72. doi: 10.1016/j.jcorpfin.2017.03.001

Krapl, A., & Giaccotto, C. (2015). Foreign exchange risk and the term-structure of industry costs

of equity. Journal of International Money and Finance, 51, 71-88. doi:

10.1016/j.jimonfin.2014.11.001

Krapl, A., & O'Brien, T. J. (2015). Direct versus indirect regression estimates of foreign

exchange cash flow exposure. International Review of Financial Analysis, 37, 103-112.

doi: 10.1016/j.irfa.2014.11.018

Kristoufek, L., & Vosvrda, M. (2016). Gold, currencies and market efficiency. Physica A:

Statistical Mechanics and its Applications, 449, 27-34. doi: 10.1016/j.physa.2015.12.075

Kuang, P., Schroeder, M., & Wang, Q. (2014). Illusory profitability of technical analysis in

emerging foreign exchange markets. International Journal of Forecasting, 30(2), 192-

205. doi: 10.1016/j.ijforecast.2013.07.015

Kudryavtsev, A., Cohen, G., & Hon-Snir, S. (2013). "Rational" or "intuitive": Are behavioral

biases correlated across stock market investors? Contemporary Economics, 7(2), 31. doi:

doi: 10.5709/ce.1897-9254.81

Kumar, D. (2013). Are PIIGS stock markets efficient? Studies in Economics and Finance, 30(3),

209-225. doi: 10.1108/SEF-09-2011-0074

Kumar, S. (2016). Revisiting calendar anomalies: Three decades of multicurrency evidence.

Journal of Economics and Business, 86, 16-32. doi: 10.1016/j.jeconbus.2016.04.001

122
Kuo, S., Chen, C., & You, C. (2017). Tick size and commonality in liquidity. Asian Economic

and Financial Review, 7(4), 431. doi: 10.18488/journal.aefr/2017.7.4/102.4.431.447

Lachanski, M., & Pav, S. (2017). Shy of the character limit: "Twitter mood predicts the stock

market" revisited. Economical Journal Watch, 14(3), 302.

Lakonishok, J., Shleifer, A., & Vishny, R. W. (1994). Contrarian investment, extrapolation, and

risk. The Journal of Finance, 49(5), 1541-1578.

Langfield, S., & Soramäki, K. (2016). Interbank exposure networks. Computational Economics,

47(1), 3-17. doi: 10.1007/s10614-014-9443-x

Lasser, D. J., & Spizman, J. D. (2016). The value of the wildcard option in cash-settled

American index options. Journal of Financial Markets, 28, 116-131. doi:

10.1016/j.finmar.2015.09.002

LeBaron, B. (1999). Technical trading rule profitability and foreign exchange intervention.

Journal of International Economics, 49, 125-143.

Lento, C., Gradojevic, N., & Wright, C. S. (2007). Investment information content in Bollinger

bands? Applied Financial Economics Letters, 3(4-6), 263-267.

Lester, B., Rocheteau, G., & Weill, P. (2015). Competing for order flow in OTC markets.

Journal of Money, Credit and Banking, 47(2), 77-126. doi: 10.1111/jmcb.12215

Levich, R. M., & Poti, V. (2015). Predictability and 'good deals' in currency markets.

International Journal of Forecasting, 31(2), 454-472. doi:

10.1016/j.ijforecast.2014.10.001

Levich, R. M., & Thomas, L. R. (1993). The significance of technical trading-rule profits in the

foreign exchange market: A bootstrap approach. Journal of International Money and

Finance, 12, 451-474.

123
Li, J. (2016). When noise trading fades, volatility rises. Review of Quantitative Finance and

Accounting, 47(3), 475-512. doi: 10.1007/s11156-015-0508-2

Lin, X., Yang, Z., & Song, Y. (2011). Intelligent stock trading system based on improved

technical analysis and Echo State Network. Expert Systems with Applications, 38(9),

11347-11354.

Liu, H., Ji, P., & Jin, J. (2016). Intra-day trading system design based on the integrated model of

wavelet de-noise and genetic programming. Entropy, 18(12), 435. doi:

10.3390/e18120435

Liu, X., An, H., Wang, L., & Guan, Q. (2017). Quantified moving average strategy of crude oil

futures market based on fuzzy logic rules and genetic algorithms. Physica A: Statistical

Mechanics and its Applications, 482, 444-457. doi: 10.1016/j.physa.2017.04.082

Lo, A. W. (2002). The statistics of Sharpe ratios. Financial Analysts Journal, 58(4), 36-52. doi:

10.2469/faj.v58.n4.2453

Lo, A. W. (2004). The adaptive markets hypothesis: Market efficiency from an evolutionary

perspective. Journal of Portfolio Management, 30(5), 15-29.

Lo, A. W. (2012). Reading about the financial crisis: A twenty-one-book review. Journal of

Economic Literature, 50(1), 151-178. doi: 10.1257/jel.50.1.151

Lubnau, T., & Todorova, N. (2015). Trading on mean-reversion in energy futures markets.

Energy Economics, 51, 312-319. doi: 10.1016/j.eneco.2015.06.018

Lustig, H., Roussanov, N., & Verdelhan, A. (2014). Countercyclical currency risk premia.

Journal of Financial Economics, 111, 527-553.

Majumder, D. (2013). Towards an efficient stock market: Empirical evidence from the Indian

market. Journal of Policy Modeling, 35(4), 572-587. doi: 10.1016/j.jpolmod.2011.08.016

124
Malkiel, B. G. (2003). The Efficient Market Hypothesis and its Critics. The Journal of Economic

Perspectives, 17(1), 59-82, doi. 10.1257/089533003321164958

Malkiel, B. G. (2005). Reflections on the Efficient Market Hypothesis: 30 Years Later. The

Financial Review, 40(1), 1-9, doi. 10.1111/j.0732-8516.2005.00090.x

Malkiel, B. G. (2015). A random walk down Wall Street: The time-tested strategy for successful

investing. New York, NY: W. W. Norton & Company.

Malkiel, B. G., & Fama, E. F. (1970). Efficient Capital Markets: A Review of Theory and

Empirical Work. The Journal of Finance, 25(2), 383-417, doi. 10.1111/j.1540-

6261.1970.tb00518.x

Manahov, V., & Hudson, R. (2014). A note on the relationship between market efficiency and

adaptability – new evidence from artificial stock markets. Expert Systems with

Applications, 41(16), 7436-7454. doi: 10.1016/j.eswa.2014.06.004

Manahov, V., Hudson, R., & Gebka, B. (2014). Does high frequency trading affect technical

analysis and market efficiency? And if so, how? Journal of International Financial

Markets, Institutions and Money, 28(1), 131-157. doi: 10.1016/j.intfin.2013.11.002

Manahov, V., Hudson, R., & Hoque, H. (2015). Return predictability and the 'wisdom of

crowds': Genetic programming trading algorithms, the marginal trader hypothesis and the

Hayek hypothesis. Journal of International Financial Markets, Institutions and Money,

37, 85-98. doi: 10.1016/j.intfin.2015.02.009

Marshall, A., Musayev, T., Pinto, H., & Tang, L. (2012). Impact of news announcements on the

foreign exchange implied volatility. Journal of International Financial Markets,

Institutions & Money, 22(4), 719-737. doi: 10.1016/j.intfin.2012.04.006

125
Marshall, B., Cahan, R., & Cahan, J. (2008a). Does intraday technical analysis in the US equity

market have value? Journal of Empirical Finance, 15(2), 199-210.

Marshall, B., Cahan, R., & Cahan, J. (2008b). Can commodity futures be profitably traded with

quantitative market timing strategies? Journal of Banking &Finance, 32(9), 1810-1819.

McCuddy, M. K., & Pirie, W. L. (2007). Spirituality, stewardship, and financial decision-

making: Toward a theory of intertemporal stewardship. Managerial Finance, 33(12),

957-969.

Mcmillan, D. G., & Speight, A. E. H. (2012). Daily FX volatility forecasts: Can the

GARCH(1,1) model be beaten using High‐Frequency data? Journal of Forecasting,

31(4), 330-343. doi: 10.1002/for.1222

Meese, R. A., & Rogoff, K. S. (1982). Empirical exchange rate models of the seventies: Do they

fit out of sample? Journal of International Economics, 14, 3-24.

Mele, A., & Sangiorgi, F. (2009). Ambiguity, information acquisition and price swings in asset

markets. FMG Discussion Paper Series No. 633.

Mende, A., & Menkhoff, L. (2006). Profits and speculation in intra-day foreign exchange

trading. Journal of Financial Markets, 9(3), 223-245. doi: 10.1016/j.finmar.2006.05.003

Menkhoff, L., Sarno, L., Schmeling, M., & Schrimpf, A. (2012). Carry trades and global foreign

exchange volatility. Journal of Finance, 67, 681-718.

Menkhoff, L., Sarno, L., Schmeling, M., & Schrimpf, A. (2016). Information flows in foreign

exchange markets: Dissecting customer currency trades. The Journal of Finance, 71(2),

601-634. doi: 10.1111/jofi.12378

Menkhoff, L., & Taylor, M. (2007). The obstinate passion of foreign exchange professionals:

Technical analysis. Journal of Economic Literature, 45(4), 936-972.

126
Meschi, E., Taymaz, E., & Vivarelli, M. (2016). Globalization, technological change and labor

demand: A firm-level analysis for turkey. Review of World Economics, 152(4), 655-680.

doi: 10.1007/s10290-016-0256-y

Metghalchi, M., Kagochi, J., & Hayes, L. A. (2014). Contrarian technical trading rules: Evidence

from Nairobi stock index. Journal of Applied Business Research, 30(3), 833.

Meyers, L. S., Gamst, G., & Guarino, A. J. (2016). Applied Multivariate Research: Design and

Interpretation. Thousand Oaks, CA: Sage.

Milionis, A. E., & Papanagiotou, E. (2013). Decomposing the predictive performance of the

moving average trading rule of technical analysis: The contribution of linear and non-

linear dependencies in stock returns. Journal of Applied Statistics, 40(11-12), 2480

-2494.

Mobarek, A., & Fiorante, A. (2014). The prospects of BRIC countries: Testing weak form

market efficiency. Research in International Business and Finance, 30, 217-232.

Mohamad, A. (2016). Short selling and the development of anti-shorting laws in the UK. Journal

of Financial Regulation and Compliance, 24(1), 9-23. doi: 10.1108/JFRC-05-2014-0025

Mollick, A. V., & Assefa, T. A. (2013). Carry-trades on the yen and the Swiss franc: Are they

different? Journal of Economics and Finance, 37, 402-423.

Morck, R. (2014). The social value of shareholder value. Corporate Governance: An

International Review, 22(3), 185-193. doi: 10.1111/corg.12063

Muijs, D. (2011). Doing quantitative research in education with SPSS (2nd ed.). Los Angeles,

CA: Sage.

Narayan, P. K. (2006). The behaviour of US stock prices: Evidence from a threshold

autoregressive model. Mathematics and Computers in Simulation, 71(2),

127
103-108. doi: 10.1016/j.matcom.2005.11.016

Narayan, P. K., Mishra, S., Narayan, S., & Thuraisamy, K. (2015). Is exchange rate trading

profitable? Journal of International Financial Markets, Institutions and Money, 38, 217-

229. doi: 10.1016/j.intfin.2015.05.015

NASDAQ. (2016, March 31). Forex market overview. Retrieved from http://www.nasdaq

.com/forex/education/foreign-exchange-market-overview.aspx

Nazário, R. T. F., Lima-eSilva, J., Amorim-Sobreiro, V., & Kimura, H. (2017). A literature

review of technical analysis on stock markets. Quarterly Review of Economics and

Finance.

Neely, C., Weller, P., & Dittmar, R. (1997). Is technical analysis in the foreign exchange market

profitable? A genetic programming approach. Journal of Financial and Quantitative

Analysis, 32(4), 405-426. doi: 10.2307/2331231

Neely, C. J., Rapach, D. E., Tun, J., & Zhou, G. (2014). Forecasting the equity risk premium:

The role of technical indicators. Management Science, 60, 1772-1791.

Neely, C. J., Weller, P. A., & Ulrich, J. (2009). The adaptive markets hypothesis: Evidence from

the foreign exchange market. Journal of Financial and Quantitative Analysis, 44, 467-

488.

Ni, Y., Liao, Y., & Huang, P. (2015). Momentum in the Chinese stock market: Evidence from

stochastic oscillator indicators. Emerging Markets Finance and Trade, 51(sup1), S99-

110. doi: 10.1080/1540496X.2014.998916

Noda, A. (2016). A test of the adaptive market hypothesis using a time-varying AR model in

Japan. Finance Research Letters, 17, 66-71. doi: 10.1016/j.frl.2016.01.004

128
Nofer, M., & Hinz, O. (2015). Using twitter to predict the stock market: Where is the mood

effect? Business & Information Systems Engineering, 57(4), 229-242. doi:

10.1007/s12599-015-0390-4

Northcott, A. (2009). The complete guide to using candlestick charting: How to earn high rates

of return-safely. Ocala: Atlantic Publishing Group Inc. No. 1601382944.

Nurunnabi, M. R (2012). Testing weak-form efficiency of emerging economies: a critical review

of literature. Journal of Business Economics & Management, 13(1), 167- 188. doi:

10.3846/16111699.2011.620140

Nwachukwu, J. C., & Shitta, O. (2015). Testing the weak-form efficiency of stock markets: A

comparative study of emerging and industrialized economies. International Journal of

Emerging Markets, 10(3), 409-426. doi: 10.1108/IJoEM-07-2013-0115

Oliveira, F. A. D., Nobre, C. N., & Zárate, L. E. (2013). Applying artificial neural networks to

prediction of stock price and improvement of the directional prediction index – Case

study of Petrobras, Brazil. Expert Systems with Applications, 40(18), 7596-7606.

Oliven, K., & Rietz, T. A. (2004). Suckers are born but markets are made: Individual rationality,

arbitrage, and market efficiency on an electronic futures market. Management Science,

50(3), 336-351. doi: 10.1287/mnsc.1040.0191

Opie, W., & Dark, J. (2015). Currency overlay for global equity portfolios: Cross‐Hedging and

base currency. Journal of Futures Markets, 35(2), 186-200. doi: 10.1002/fut.21662

Orlov, V. (2016). Currency momentum, carry trade, and market illiquidity. Journal of Banking

and Finance, 67, 1-11. doi: 10.1016/j.jbankfin.2016.02.010

Osiyevskyy, O., & Biloshapka, V. (2017). Shareholders: Managing them as allies, partners and

significant constituents. Strategy & Leadership, 45(5), 41.

129
Ozturk, M., Toroslu, I. H., & Fidan, G. (2016). Heuristic based trading system on forex data

using technical indicator rules. Applied Soft Computing, 43, 170-186. doi:

10.1016/j.asoc.2016.01.048

Pan, L., Tang, Y., & Xu, J. (2016). Speculative trading and stock returns. Review of Finance,

20(5), 1835-1865. doi: 10.1093/rof/rfv059

Papadamou, S., & Markopoulos, T. (2014). Investigating intraday interdependence between

gold, silver and three major currencies: The Euro, British Pound and Japanese Yen.

International Advances in Economic Research, 20(4), 399-410. doi: 10.1007

/s11294-014-9490-z

Parylo, O. (2012). Qualitative, quantitative, or mixed methods: An analysis of research design in

articles on principal professional development (1998-2008). International Journal of

Multiple Research Approaches, 6(3), 297-313.

Phillips, P. C. B., & Yu, J. (2011). Dating the timeline of financial bubbles during the subprime

crisis. Quantitative Economics, 2(3), 455-491.

Pintar, R., Jereb, E., Cudanov, M., & Urh, M. (2016). Interest in currency trading learning -

preferred methods and motivational factors. Organizacija, 49(1), 3-14. doi: 10.1515/orga-

2016-0003

Plott, C. R., & Pogorelskiy, K. (2017). Call market experiments: Efficiency and price discovery

through multiple calls and emergent newton adjustments. American Economic Journal.

Microeconomics, 9(4), 1. doi: 10.1257/mic.20150201

Popovic, S., & Durovic, A. (2014). Intraweek and intraday trade anomalies: evidence from

FOREX market. Applied Economics, 46(32), 3968-3979.

130
Poti, V., Levich, R. M., Pattitoni, P., & Cucurachi, P. (2014). Predictability, trading rule

profitability and learning in currency markets. International Review of Financial

Analysis, 33, 117-129. doi: 10.1016/j.irfa.2014.01.001

Poti, V., & Siddique, A. (2013). What drives currency predictability? Journal of International

Money and Finance, 36, 86-106. doi: 10.1016/j.jimonfin.2013.03.004

Powers, C. (2016). Filling the regulatory void in the FX spot market: How traders rigged the

biggest market in the world. Fordham Urban Law Journal, 43(1), 167-179.

Prokopczuk, M., Symeonidis, L., & Wese Simen, C. (2016). Do jumps matter for volatility

forecasting? Evidence from energy markets. Journal of Futures Markets, 36(8), 758-792.

doi: 10.1002/fut.21759

Pukthuanthong, K., Levich, R. M., & Thomas, L. R. (2007). Do foreign exchange markets still

trend? Journal of Portfolio Management, 34(1), 114-128.

Raj, M. (2013). The use of moving average trading rules to generate abnormal positive returns in

the intra-day futures market. International Journal of Management, 30(3), 36-42.

Ranaldo, A. (2009). Segmentation and time-of-day patterns in foreign exchange markets.

Journal of Banking & Finance, 33, 2199-2206.

Rime, D., Sarno, L., & Sojli, E. (2010). Exchange rate forecasting, order flow, and

macroeconomic information. Journal of International Economics, 80, 72-88.

Rime, D., & Schrimpf, A. (2013). The anatomy of the global FX market through the lens of the

2013 Trienniel survey. Bank of International Settlements, 19, 29-46.

Rogoff, K., & Stavrakeva, V. (2008). The continuing puzzle of short horizon exchange rate

forecasting. NBER Working Paper No. 14071.

131
Romeo, A., Joseph, G., & Elizabeth, D. T. (2015). A study on the formation of candlestick

patterns with reference to nifty index for the past five years. International Journal of

Management Research and Reviews, 5(2), 67.

Rosch, D. M., Subrahmanyam, A., & van Dijk, M. A. (2017). The dynamics of market

efficiency. Review of Financial Studies, 30(4), 1151-1187.

Rudestam, K. E., & Newton, R. R. (2015). Surviving your dissertation (4th ed.). Thousand Oaks,

CA: Sage.

Rzepczynski, M. S. (2008). Foreign exchange: A practical guide to the FX markets. Financial

Analysts Journal, 64(3), 99-100.

Sager, M. J., & Taylor, M. P. (2006). Under the microscope: The structure of the foreign

exchange market. International Journal of Finance & Economics, 11(1), 81-95. doi:

10.1002/ijfe.277

Sarno, L., & Schmeling, M. (2014). Which fundamentals drive exchange rates? A Cross‐

Sectional perspective. Journal of Money, Credit and Banking, 46(2-3), 267-292. doi:

10.1111/jmcb.12106

Satterlee, B. C. (2014). Cross border commerce: With biblical worldview application (2nd ed.).

Raleigh, NC: SI-CORP.

Savin, G., Weller, P., & Zvingelis, J. (2007). The predictive power of "head-and-shoulders" price

patterns in the U.S. stock market. Journal of Financial Econometrics, 5(2), 243-265. doi:

10.1093/jjfinec/nbl012

Schulmeister, S. (2006). The interaction between technical currency trading and exchange rate

fluctuations. Finance Research Letters, 3, 212-233.

132
Schulmeister, S. (2009). Profitability of technical stock trading: Has it moved from daily to

intraday data? Review of Financial Economics, 18(4), 190-201. doi:

10.1016/j.rfe.2008.10.001

Serbinenko, A., & Rachev, S. T. (2010). A new hybrid model for intraday spot foreign exchange

trading accounting for heavy tails and volatility clustering. Journal of Computational

Analysis and Applications, 12, 337-360.

Shalini, K. S. (2012). Neurofinance and investment behaviour. Studies in Economics and

Finance, 29(4), 246-267. doi: 10.1108/10867371211266900

Shen, X., & Hartarska, V. (2013). Derivatives as risk management and performance of

agricultural banks. Agricultural Finance Review, 73(2), 290-309. doi: 10.1108/AFR-07-

2012-0036

Shynkevich, A. (2017). Two tales of return predictability: The case of Asia–Pacific equity

markets. Journal of Forecasting, 36(3), 257-272. doi: 10.1002/for.2436

Singh, R., & Srivastava, S. (2017). Stock prediction using deep learning. Multimedia Tools and

Applications, 76(18), 18569-18584. doi: 10.1007/s11042-016-4159-7

Siow, G. (2013). Transaction costs and suitability of trading currency - Case on selected East

Asian economies. International Journal of Trade: Economics and Finance, 4(1), 25-32.

doi: 10.7763/IJTEF.2013.V4.255

Sobaci, C., Sensoy, A., & Erturk, M. (2014). Impact of short selling activity on market

dynamics: Evidence from an emerging market. Journal of Financial Stability, 15, 53-62.

doi: 10.1016/j.jfs.2014.08.010

Sobreiro, V. A., da Costa, T. R. C. C., Nazario, R. T. F., e Silva, J. L., Moreira, E. A., & Lima

Filho, M. C. … & Zambrano, J. C. A. (2016). The profitability of moving average trading

133
rules in BRICS and emerging stock markets. North American Journal of Economics and

Finance, 38, 86-101. doi: 10.1016/j.najef.2016.08.003

Song, S. (2015). Inter-country exchange rates and intra-firm trade flow within global network of

multinational corporations. Management International Review, 55(1), 1-22. doi:

10.1007/s11575-014-0233-4

Spector, J. M., Merrill, M. D., Elen, J., & Bishop, M. J. (2013). Handbook of research for

educational communication and technology (4th ed.). New York, NY: Springer.

Stake, R. E. (2010). Qualitative research: Studying how things work. London, UK: Gilford Press.

Stefanescu, R., & Dumitriu, R. (2016). Contrarian and momentum profits during periods of high

trading volume preceded by stock prices shocks. Risk in Contemporary Economy, 3, 378-

384.

Stoll, H. R. (1973). The relationship between put and call option prices. The Journal of Finance,

28(1), 185-187.

Sullivan, R., Timmermann, A., & White, H. (1999). Data-snooping, technical trading rule

performance, and the bootstrap. Journal of Finance, 54(5), 1647-1691.

Sweeney, R. J. (1986). Beating the foreign exchange market. Journal of Finance, 41, 163-182.

Szakmary, A. C., Shen, Q., & Sharma, S.C. (2010). Trending-following trading strategies in

commodity futures: a re-examination. Journal of Banking & Finance, 34, 409-426.

Talebi, H., Winsor, H., & Gavrilova, M. L. (2014). Multi-scale foreign exchange rates ensemble

for classification of trends in forex market. Procedia Computer Science, 29, 2065-2075.

doi: 10.1016/j.procs.2014.05.190

Taylor, C. S. (2014). Validity and validation. Oxford, UK: Oxford University Press. doi:

10.1093/acprof:osobl/9780199791040.001.0001

134
Temnov, G. (2017). A strategy based on mean reverting property of markets and applications to

foreign exchange trading with trailing stops. Applied Stochastic Models in Business and

Industry, 33(2), 152-166. doi: 10.1002/asmb.2229

Thinyane, H., & Millin, J. (2011). An investigation into the use of intelligent systems for

currency trading. Computational Economics, 37(4), 363-374. doi: 10.1007/s10614-011-

9260-4

Thompson, R. (1978). The information content of discounts and premiums on closed-end fund

shares. Journal of Financial Economics, 6(2), 151-186. doi: 10.1016/0304-

405X(78)90028-4

TickData, LLC. (2017, December). Forex FAQ. Retrieved from https://www.tickdata.com/forex-

faq/#pane346

Timmermann, A., & Granger, C. W. J. (2004). Efficient market hypothesis and forecasting.

International Journal of Forecasting, 20(1), 15-27. doi: 10.1016/S0169-2070(03)00012-8

Tsang, C. (2017). Rethinking modern financial ecology and its regulatory implications. Banking

& Finance Law Review, 32(3), 461-494.

Van Duzer, J. B. (2010). Why business matters to God: And what still needs to be fixed. Downers

Grove, IL: InterVarsity Press.

Vanstone, B., & Finnie, G. (2009). An empirical methodology for developing stock market

trading systems using artificial neural networks. Expert Systems with Applications, 36(3),

6668-6680.

Velez, O. L., & Capra, G. (2000). Tools and tactics for the master day trader: Battle-tested

techniques for day, swing, and position traders. New York, NY: McGraw-Hill.

135
Verheyden, T., Van den Bossche, F., & De Moor, L. (2015). Towards a new framework on

efficient markets. Research in International Business and Finance, 34, 294-308.

Vives, X. (2014). On the possibility of informationally efficient markets. Journal of the

European Economic Association, 12(5), 1200-1239. doi: 10.1111/jeea.12107

Wang, X., Smith, K. A., & Hyndman, R. J. (2005). Dimension reduction for clustering time

series using global characteristics. Proceedings of the 5th International Conference on

Computational Science, 3516, 792-795. doi: 10.1007/11428862_108

Westerlund, J., & Narayan, P. (2013). Testing the efficient market hypothesis in conditionally

heteroskedastic futures markets. Journal of Futures Markets, 33(11), 1024-1045. doi:

10.1002/fut.21624

Westerlund, J., Norkute, M., & Narayan, P. K. (2015). A factor analytical approach to the

efficient futures market hypothesis. Journal of Futures Markets, 35(4), 357-370. doi:

10.1002/fut.21687

Williams, O. D. (2006). Empirical optimization of Bollinger Bands for profitability. Applied

Economic Letters, 2006. doi: 10.2139/ssrn.2321140

Wong, A. H., & Heaney, R. A. (2017). Volatility smile and One‐Month foreign currency

volatility forecasts. Journal of Futures Markets, 37(3), 286-312. doi: 10.1002/fut.21799

Wong, K. P. (2016). Ambiguity and the multinational firm. International Review of Economics

and Finance, 43, 404-414. doi: 10.1016/j.iref.2016.01.004

Wright, J. H. (2012). What does monetary policy do to long‐term interest rates at the zero lower

bound? The Economic Journal, 122(564), F447-F466. doi: 10.1111/j.1468-

0297.2012.02556.x

136
Xu, S., & Yang, Y. (2013). Fractional Black-Scholes model and technical analysis of stock price.

Journal of Applied Mathematics, 1-7. doi: 10.1155/2013/631795

Yan, X. X., Zhang, Y. B., Lv, X. K., & Li, Z. Y. (2017). Improvement and test of stock index

futures trading model based on Bollinger Bands. International Journal of Economics and

Finance, 9(1), 78-87.

Yin, R. K. (2006). Mixed methods research: Are the methods genuinely integrated or merely

parallel?. Research in the Schools, 13(1), 41-47.

Yin, R. K. (2011). Qualitative research from start to finish (2nd ed.). London, UK: Gilford Press.

Zakamulin, V. (2014). The real-life performance of market timing with moving average and

time-series momentum rules. Journal of Asset Management, 15(4), 261-278. doi:

10.1057/jam.2014.25

Zarrabi, N., Snaith, S., & Coakley, J. (2017). FX technical trading rules can be profitable

sometimes. International Review of Financial Analysis, 49, 113-127. doi:

10.1016/j.irfa.2016.12.010

Zhou, J., & Lee, J. M. (2013). Adaptive market hypothesis: Evidence from the REIT market.

Applied Financial Economics, 23(21), 1649-1662.

Zhu, H., Jiang, Z. Q., Li, S. P., & Zhou, W. X. (2015). Profitability of simple technical trading

rules of Chinese stock exchange indexes. Physica A: Statistical Mechanics and its

Applications, 439(1), 75-84.

Zhu, Y., & Zhou, G. (2009). Technical analysis: An asset allocation perspective on the use of

moving averages. Journal of Financial Economics, 92(3), 519-544.

Zoicas-Ienciu, A. (2016). How technical is technical analysis? Review of Economic Studies and

Research, 9(2), 155-171.

137
Appendix A: Listing of all Days

2009 2010 2011

Monday, January 5, 2009 Tuesday, January 5, 2010 Wednesday, January 5, 2011
Friday, January 9, 2009 Tuesday, January 12, 2010 Tuesday, February 8, 2011
Friday, January 30, 2009 Wednesday, February 3, 2010 Wednesday, March 2, 2011
Monday, March 16, 2009 Friday, February 5, 2010 Friday, March 4, 2011
Thursday, April 2, 2009 Friday, March 5, 2010 Friday, March 25, 2011
Tuesday, April 14, 2009 Wednesday, April 7, 2010 Monday, April 11, 2011
Monday, April 20, 2009 Tuesday, June 8, 2010 Wednesday, April 13, 2011
Thursday, May 14, 2009 Monday, June 14, 2010 Thursday, May 12, 2011
Monday, May 25, 2009 Tuesday, July 27, 2010 Friday, June 10, 2011
Friday, June 12, 2009 Tuesday, August 17, 2010 Monday, June 20, 2011
Tuesday, June 30, 2009 Monday, August 30, 2010 Friday, July 15, 2011
Friday, July 31, 2009 Thursday, September 2, 2010 Monday, July 18, 2011
Wednesday, September 9, 2009 Thursday, September 23, 2010 Monday, July 25, 2011
Monday, September 14, 2009 Tuesday, September 28, 2010 Thursday, August 11, 2011
Friday, September 18, 2009 Monday, October 18, 2010 Friday, September 23, 2011
Monday, September 28, 2009 Wednesday, October 20, 2010 Wednesday, September 28, 2011
Monday, October 26, 2009 Thursday, October 21, 2010 Tuesday, October 18, 2011
Thursday, November 5, 2009 Thursday, November 11, 2010 Thursday, October 20, 2011
Tuesday, November 10, 2009 Tuesday, November 30, 2010 Friday, October 21, 2011
Thursday, December 31, 2009 Thursday, December 16, 2010 Thursday, December 22, 2011

2012 2013
Monday, January 16, 2012 Monday, March 4, 2013
Monday, February 6, 2012 Monday, April 22, 2013
Monday, February 13, 2012 Wednesday, May 1, 2013
Monday, February 20, 2012 Friday, May 3, 2013
Tuesday, February 21, 2012 Thursday, May 9, 2013
Friday, February 24, 2012 Friday, May 10, 2013
Friday, March 23, 2012 Monday, May 13, 2013
Friday, April 6, 2012 Monday, May 20, 2013
Thursday, June 28, 2012 Monday, July 8, 2013
Friday, July 6, 2012 Thursday, July 18, 2013
Wednesday, July 25, 2012 Wednesday, July 31, 2013
Monday, August 6, 2012 Thursday, August 1, 2013
Monday, August 13, 2012 Thursday, August 22, 2013
Tuesday, September 11, 2012 Friday, August 30, 2013
Tuesday, September 25, 2012 Thursday, September 19, 2013
Friday, October 5, 2012 Wednesday, September 25, 2013
Monday, October 8, 2012 Thursday, September 26, 2013
Wednesday, October 10, 2012 Monday, October 21, 2013
Tuesday, October 16, 2012 Thursday, November 14, 2013
Friday, December 28, 2012 Monday, November 18, 2013

138
2014 2015
Wednesday, January 1, 2014 Wednesday, January 7, 2015
Monday, January 20, 2014 Wednesday, February 18, 2015
Wednesday, February 12, 2014 Tuesday, March 17, 2015
Friday, March 7, 2014 Tuesday, March 24, 2015
Tuesday, March 18, 2014 Tuesday, April 14, 2015
Monday, March 31, 2014 Thursday, May 7, 2015
Thursday, April 17, 2014 Friday, May 8, 2015
Tuesday, April 29, 2014 Friday, May 15, 2015
Tuesday, May 6, 2014 Friday, July 17, 2015
Monday, May 12, 2014 Tuesday, July 21, 2015
Tuesday, May 20, 2014 Friday, July 24, 2015
Thursday, May 22, 2014 Friday, July 31, 2015
Monday, June 2, 2014 Friday, August 7, 2015
Wednesday, June 4, 2014 Tuesday, September 1, 2015
Monday, June 23, 2014 Monday, September 7, 2015
Thursday, July 17, 2014 Thursday, October 22, 2015
Friday, September 12, 2014 Monday, November 2, 2015
Thursday, September 25, 2014 Thursday, November 12, 2015
Tuesday, December 9, 2014 Monday, November 23, 2015
Thursday, December 11, 2014 Monday, December 28, 2015

2016
Wednesday, January 6, 2016
Friday, February 12, 2016
Tuesday, March 1, 2016
Tuesday, March 29, 2016
Friday, May 27, 2016
Monday, May 30, 2016
Wednesday, June 22, 2016
Wednesday, July 13, 2016
Tuesday, July 26, 2016
Friday, July 29, 2016
Tuesday, August 2, 2016
Thursday, August 18, 2016
Wednesday, August 31, 2016
Monday, October 24, 2016
Thursday, October 27, 2016
Thursday, November 3, 2016
Monday, November 14, 2016
Monday, November 21, 2016
Friday, December 2, 2016
Wednesday, December 14, 2016

139
Appendix B: Normality Test by Year Using Alpha of 0.05

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 1.41946 5.99146 0.49178 TRUE
2009
Shapiro-Wilk 0.96694 #N/A 0.68948 TRUE
Doornik Chi-Square 2.42747 5.99146 0.29709 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 0.63966 5.99146 0.72627 TRUE
2010
Shapiro-Wilk 0.97458 #N/A 0.84710 TRUE
Doornik Chi-Square 0.72405 5.99146 0.69627 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 1.86062 5.99146 0.39443 TRUE
2011
Shapiro-Wilk 0.94083 #N/A 0.24854 TRUE
Doornik Chi-Square 3.23364 5.99146 0.19853 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 1.80743 5.99146 0.40506 TRUE
2012
Shapiro-Wilk 0.92360 #N/A 0.11626 TRUE
Doornik Chi-Square 4.33542 5.99146 0.11444 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 1.31182 5.99146 0.51897 TRUE
2013
Shapiro-Wilk 0.96419 #N/A 0.63051 TRUE
Doornik Chi-Square 2.02797 5.99146 0.36277 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 0.36306 5.99146 0.83399 TRUE
2014
Shapiro-Wilk 0.96302 #N/A 0.60568 TRUE
Doornik Chi-Square 0.54164 5.99146 0.76275 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 1.77657 5.99146 0.41136 TRUE
2015
Shapiro-Wilk 0.93983 #N/A 0.23806 TRUE
Doornik Chi-Square 2.76296 5.99146 0.25121 TRUE

Normality Test Score C.V. P-Value Pass?

Jarque-Bera 1.40362 5.99146 0.49569 TRUE
2016
Shapiro-Wilk 0.96343 #N/A 0.61431 TRUE
Doornik Chi-Square 2.52524 5.99146 0.28291 TRUE

140
Appendix C: ANOVA by Year

2009 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 17330 866.5 3540434.474
Long (Net gain or Loss) 20 -9540 -477 1765264.211
Short (Net Gain or Loss) 20 8250 412.5 1768840.789

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 18682123.3 2 9341061.667 3.961132043 0.02450271 3.158843
Within Groups 134416250 57 2358179.825
Total 153098373 59

2010 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 2130 106.5 462013.4211
Long (Net gain or Loss) 20 -1830 -91.5 764550.2632
Short (Net Gain or Loss) 20 540 27 766232.6316

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 397110 2 198555 0.298909123 0.742782126 3.15884
Within Groups 37863130 57 664265.439
Total 38260240 59

2011 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 -2230 -111.5 594855.5263
Long (Net gain or Loss) 20 -760 -38 521427.3684
Short (Net Gain or Loss) 20 -390 -19.5 514310.2632

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 94723.33333 2 47361.66667 0.087137002 0.91667337 3.15884
Within Groups 30981270 57 543531.0526
Total 31075993.33 59

2012 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 -500 -25 208531.5789
Long (Net gain or Loss) 20 2000 100 327084.2105
Short (Net Gain or Loss) 20 -3100 -155 331805.2632

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 650333.333 2 325166.6667 1.124598022 0.33188164 3.15884
Within Groups 16481000 57 289140.3509
Total 17131333.3 59

141
2013 Anova: Single Factor
SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 -2710 -135.5 406541.8421
Long (Net gain or Loss) 20 -1110 -55.5 234057.6316
Short (Net Gain or Loss) 20 240 12 239006.3158

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 218083.3333 2 109041.6667 0.371899553 0.69108395 3.15884272
Within Groups 16712510 57 293201.9298
Total 16930593.33 59

2014 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 648 32.4 58219.93684
Long (Net gain or Loss) 20 -410 -20.5 107415.5263
Short (Net Gain or Loss) 20 -300 -15 107794.7368

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 33836.13333 2 16918.0667 0.185620316 0.831088946 3.15884
Within Groups 5195173.8 57 91143.4
Total 5229009.933 59

2015 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 -132426 -6621.3 12749537.06
Long (Net gain or Loss) 20 -260 -13 576022.1053
Short (Net Gain or Loss) 20 -460 -23 578032.6316

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 581381945 2 290690972.6 62.72285107 3.98222E-15 3.158843
Within Groups 264168244 57 4634530.6
Total 845550189 59

2016 Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Bollinger Bands (Net Gain or Lo 20 -62820 -3141 17865146.32
Long (Net gain or Loss) 20 1380 69 343230.5263
Short (Net Gain or Loss) 20 -2080 -104 337193.6842

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 130382653.3 2 65191326.67 10.54558983 0.00012685 3.15884
Within Groups 352365840 57 6181856.842
Total 482748493.3 59

142
Appendix D: t-Test when ANOVA was Significant

t-Test: 2015 t-Test: 2015 t-Test: 2015

Bollinger Bands Buy and Hold Bollinger Bands Sell and Hold Buy and Hold Sell and Hold
Mean -6621.3 -13 Mean -6621.3 -23 Mean -13 -23
Variance 12749537.06 576022.1053 Variance 12749537.06 578032.6316 Variance 576022.1053 578032.6316
Observations 20 20 Observations 20 20 Observations 20 20
Pooled Variance 6662779.584 Pooled Variance 6663784.847 Pooled Variance 577027.3684
Hypothesized Diff. 0 Hypothesized Diff. 0 Hypothesized Diff. 0
df 38 df 38 df 38
t Stat -8.095842068 t Stat -8.082981301 t Stat 0.041629557
P(T<=t) one-tail 4.26568E-10 P(T<=t) one-tail 4.43426E-10 P(T<=t) one-tail 0.48350599
t Critical one-tail 1.68595446 t Critical one-tail 1.68595446 t Critical one-tail 1.68595446
P(T<=t) two-tail 8.53135E-10 P(T<=t) two-tail 8.86852E-10 P(T<=t) two-tail 0.967011979
t Critical two-tail 2.024394164 t Critical two-tail 2.024394164 t Critical two-tail 2.024394164

t-Test: 2016 t-Test: 2016 t-Test: 2016

Bollinger Bands Buy and Hold Bollinger Bands Sell and Hold Buy and Hold Sell and Hold
Mean -3141 69 Mean -3141 -104 Mean 69 -104
Variance 17865146.32 343230.5263 Variance 17865146.32 337193.6842 Variance 343230.5263 337193.6842
Observations 20 20 Observations 20 20 Observations 20 20
Pooled Variance 9104188.421 Pooled Variance 9101170 Pooled Variance 340212.1053
Hypothesized Diff. 0 Hypothesized Diff. 0 Hypothesized Diff. 0
df 38 df 38 df 38
t Stat -3.364220197 t Stat -3.183436406 t Stat 0.937931716
P(T<=t) one-tail 0.000882082 P(T<=t) one-tail 0.00145073 P(T<=t) one-tail 0.177101501
t Critical one-tail 1.68595446 t Critical one-tail 1.68595446 t Critical one-tail 1.68595446
P(T<=t) two-tail 0.001764164 P(T<=t) two-tail 0.002901461 P(T<=t) two-tail 0.354203001
t Critical two-tail 2.024394164 t Critical two-tail 2.024394164 t Critical two-tail 2.024394164

143