Chapter 08

Regression Analysis

Afjal Hossain
Assistant Professor
Department of Marketing, PSTU

R egression A nalysis
Regression analysis is a branch of statistical

theory used to measure or estimate the

associative relationship between dependent
and independent variable.
Ex: (01) Two variables; price (X) and demand

(Y) are closely related. So, from these two

variables we can find out the most probable
value of X for a given value of Y and viceversa.
Ex: (02) Height of Fathers and Sons

A pplications of R egression A nalysis

1) It

provides estimates of values of the

dependent
variables
from
values
of
independent variables. The device used to
accomplish the estimation procedure is the
regression line which describes the average
relationship existing between X and Y
variables. It is also known as estimating line/
predicting line.
2) To obtain a measure of the error involved in
using the regression line as a basis for
estimation. If little scatter is observed then it
is said that good estimate is made and viceversa.
3) It is possible to obtain a measure of the

The Linear B ivariate R egression

M odel
Bivariate

Regression is a procedure for

deriving a mathematical relationship, in the
form of an equation, between a single metric
dependent or criterion variable and a single
metric independent or predictor variable.

Y a bX
Model:

A pplications of B ivariate R egression

Variation in sales in terms of variation in

advertising expenditures
Variation in market share in terms of
variation in sales force
Consumers perception of quality in terms of
perception of price

R egression Equation
Regression Equation of YYonX:
a bX
e
To determine or calculate the values of a & b the

following two normal equations are to be solved

1. Y Na b X
simultaneously:
2. XY a X X

where,

Ye Dependent variable to be estimated

X Independent variable
a & b Unknown constants (Fixed numerical values)

Sometimes a & b are called as the parameters of a

regression line. If the value of either or both of them is

changed, another line is determined. The parameter a
determines the level of the fitted line (distance of the line
directly above or below the origin). The parameter b

R egression Equation
Regression Equation of XXon
aY: bY
To determine or calculate the values of a & b,

the following two normal equations are to be

simultaneously:
1solved
. X Na
b Y

2. XY a Y Y

R egression C o-ef c
i
fient
Regression Co-efficient: The quantity b in the

regression equations is called the regression

co-efficient. It is sometimes called slope coefficient. Since there are two regression
equations,
therefore,
there
are
two
regressions co-efficient are:

1. Regression Co - efficient of X on Y, b xy r

2. Regression Co - efficient of Y on X, b yx r

y
y
x

P roperties of R egression C o-ef c

i
fient
The co-efficient of correlation is the geometric mean of the

r b xy x b yx
two regressions co-efficients. Symbolically,

If one of the regression co-efficients is greater that unity, the

other must be less than unity, since the value of the coefficient or correlation cannot exceed unity.
Both the regression co-efficients will have the same sign, i.e.

they will be either positive or negative. In other word, it is

not possible that one of the regression coefficients is having
minus sign and the other plus sign.
The co-efficient of correlation will have the same sign as that

of regression co-efficients, i.e. if regression co-efficients have

a negative sign, r will also have negative sign and if the
regression co-efficients have a positive sign, r would also be
positive.
The average value of the two regression co-efficients would

be greater than the value of co-efficients of correlation.

Regression co-efficients are independent of change of origin

but not scale.

P roperties of R egression C o-ef c

i
fient
Standard error of Estimate: The measure of variation of
the observations around the computed regression line is
referred to as the standard error of the estimate. It is also
Y a Y XY
denoted as SEE.

1. SEE of regression equation of y on x : b yx

2. SEE of regression equation of x on y : b xy

X a X XY

N-2

N-2

The Standard error of estimate can very easily be calculated

with the help of the following formula:

1. SEE of regression equation of y on x : S yx Sx 1 r

2. SEE of regression equation of x on y : Sxy S y 1 r

P roperties of R egression C o-ef c

i
fient
The standard error of estimate measures the

accuracy of the estimated figures. The

smaller the value of standard error of
estimate, the closer will be dots to the
regression line and the better the estimates
based on the equation for this line. If
standard error of estimate is zero, then there
is no variation about the line and the
correlation will be perfect. Thus, with the help
of standard error of estimate it is possible for
us to ascertain how good and representative
the regression line is as a description of the
average relationship between two series.

C o-ef c
i
fient of D eterm ination
The ratio of the unexplained variation to the

total variation represents the proportion of

variation in Y that is not explained by
regression on X. Subtraction of this
proportion from 1 gives the proportion of
variation in Y that is explained by regression
on X. the statistic usedR 2 to express this
Variation in
regression
on X
proportion
is Y remaining
called after
the
coefficient
of
R 2 1
Total is
variation
in Y by
determination and
denoted
whereR 2 1

Error Sum of Squares

Total Sum of Squares
R2

The value of

variation

in

is the proportion of the

the dependent variable Y

Exercises: