Scribd
Upload
133 views12 pages

Multiple Linear Regression

In this Presentation I was explain about hoe we can did MLR using Microsoft Excel.

Uploaded by

Smart academy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
133 views12 pages

Multiple Linear Regression

In this Presentation I was explain about hoe we can did MLR using Microsoft Excel.

Uploaded by

Smart academy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

MLR

PRESENTED BY : Jay S Rathod [ R19MTE14 ]


SUBJECT : Urban Transport Planning
M.Tech in Transportation Engineering
SCHOOL OF CIVIL ENGINEERING

1
CONTENT :

Introduction
MLR Equation
Assumption in MLR Analysis
Validity of MLR Analysis
Aggregated Analysis
Disaggregated Analysis
Criteria for evaluation of Regression Equation
Advantages & Disadvantages of MLR
References 2
INTRODUCTION :

MLR = Multiple Linear Regression


Multiple linear regression analysis is a well-known statistical technique for
fitting mathematical relationships between dependant and independent
variable.
This technique has become a very powerful tool in the hands of the
transportation planner.

3
MLR Equation :

In trip generation equation , the dependent variable is the number of trips and the
independent variables are the various measurable factor that influence trip generation.

Yp = a1X1 + a2X2 + a3X3 ,...., anXn + U


Where, Yp = number of trips for specified purpose p
X1,X2,X3,......,Xn = independent variables relating to
a1,a2,a3,......,an = coefficients of the respective independent variable.
U = Disturbance [Constant]
The above equation is used for determining the future values of trips, knowing the
estimated future values of the independent variables.
4
MLR Equation :

Examples:
The following equation developed from a study of Toronto;
Total trip = (0.318)x(population 5 year and older)+(0.458)x(no of household)
+ (0.890)x(no of cars owned)
The another example below is from a study of Modesto;
Y=2.18+3.404A+0.516H+0.0119 X1 – 0.343 X2
Where; Y= Average trips per occupied dwelling unit
A= Car ownership H= Household size
X1 = Social rank index X2 = Urbanization index
5
Assumptions in MLR Analysis :

All the variables are independent of each other


All the variables are normally distributed
All the variables are continuous
A linear relationship exists between the dependent variable and the
independent variable.
Influence of independent variable is additive, that is the inclusion of each
variable in the equation contributes a distinct portion of the trip numbers.

6
Validity of MLR Analysis :

Independent variables in the regression equation are not truely independent


of each other, and some sort of correlation normally exists among them.
Variable such as car ownership, family income, residential density etc are
all interrelated to a certain extent.
Many of these variables are not “Normally-Distributed”.
Some of them are not continuous variables.

7
Aggregated Analysis:

It is also known as Zonal Least-Square regression.


In this analysis each traffic zone is treated as one observation.
It is most widely used analysis method.
It is based on the assumption that contiguous household exhibit a certain
amount of similarity in travel characteristics.
This assumption allows the data in a zone to be grouped and the mean value
of the independent variable used in further calculations.

8
Disaggregated Analysis:

It is also known as Household least-square regression.


In this analysis each household is treated as an observation.
It is not so widely used.
In this process, all the enormous amount of data is used more effectively
resulting in a more meaningful description of the characteristics.
As compared to aggregated analysis, disaggregated analysis produces better
results.
It is considered more likely to be stable over time and to provide more
reliable future estimates. 9
Criteria for Evaluation of Regression Equation:

The multiple correlation coefficient should have a value at least 0.75 or


even higher. A value close to 1.0 shows a very good correlation.
The standard error of the estimate of the dependent variable should be
sufficiently small.
The F test should be carried out to examine evidence of the degree of
certainty that a meaningful relationship exists between the dependent and
independent variables.
The equation should have accuracy, validity, simplicity, sharpness and
constancy.

10
Advantages & Disadvantages of MLR:
Advantages of MLR:
Linear equation is easy to understand and derive.
Less data is required.

Disadvantages of MLR:
The equation derived is purely empirical in nature and fails to establish a
meaningful relationship between the dependant and independent variables.
The technique is based on the premise that the regression coefficients
initially established will still remain unchanged in the future and can be
used in the regression equation for predicting future travel.

11
REFERANCES :

Dr.L.R.Kadiyali publish a book on “Traffic Engineering and Transport


Planning” at Khanna Publication, pg. No. 617 – 620.

Alaa Mohammad Yousef Dodeen make a report on “Developing Trip


Generation Models Utilizing Linear Regression Analysis: Jericho City as a
Case Study” An- Najah National University Faculty of Graduate Studies,
pg. No. 36 - 44.

12

You might also like