17-1

Chapter 7: Regression
 17-2

Meaning
■ Regression is the measure of average
relationship between two or more
variables in terms of the original units of
the data.
■ Regression analysis is a statistical tool to
study the nature and extent of functional
relationship between two or more
variables.
 17-3

Steps
■ IfCorrelation is established among given
variables .
■ Then find out the average relationship
between the two .
■ To estimate the unknown values of
dependent variables from the known
values of independent variables.
 17-4

Regression Equation:-

A mathematical Equation that allow us to

predict the value of one variable from known
values of one or more other variables.
■ Dependent Variables- (explained variables,
response or outcome variable (y))-The variables whose
value is predicted on the basis of another one.
■ Independent Variables- (explanatory variables,
predictor or input variable (x))-The variables which are
used to predict the value of a dependent variable.
 17-5

Example:-
When sales are predicted on the basis of
advertising expenditure ,
sale is- ‘dependent variable’
advertising expenditure is- ‘independent
variable’.
 17-6

Uses
■ In all fields
■ Regression analysis predicts the value of
dependent variables from the value of
independent variables.
■ used to calculate r² and r .
■ Regression analysis is used in statistical
estimation of demand curves, supply curves,
production function, cost function and
consumption function.
 17-7

■ Types of regression
o Simple Linear Regression
o Multiple Regression
o Nonlinear Regression
■ Simple linear regression aims to find a linear
relationship between a response variable and a
possible predictor variable by the method of least
squares.
 17-8

Regression
■ Simple linear regression analysis estimates
the relationship between two variables.
■ Simple linear regression estimates
relationships of the form y = a + bx.
■ While the terminology is such that we say
that X "predicts" Y, we cannot say that X
"causes" Y.
 17-9

■ Regression line of X on Y and Y on X

D
E
P
E
N
D X on Y
E
N
t
V Y on X
A
R
I
A Independent variable
B
L
e
 17-10

X-axis Y-axis
■ The names of the
variables on the X and
independent dependent
Y axis vary according to
the field of application.
predictor predicted
■ The analysis is always
described as the
regression of the carrier response

response on the carrier

input output
■ Y = a + bX
 17-11

The Regression Equation

■ A linear regression equation is usually written
Y = a + bX
where
Y is the dependent variable
a is the constant (intercept)
b is the regression coefficient (slope)
X is the independent variable (or covariate)

*here we are interested in finding the values of a and b.

 17-12

■ The most commonly used technique for fitting a

straight line to a scatter diagram is the least-squares
procedure.

■ In fitting the line, the least-squares procedure

minimizes the sum of squared errors
or
■ The line of best fit is the one which minimizes the
sum of the squares of the vertical distances from the
observed points to the line.
 17-13

Regression equation :
■ Regression equation of X on Y:
Xe = a + bY
where
Xe : dependent variable
Y : independent variable
a & b : constants to be estimated

“It measures the amount of change in X

corresponding to unit change in Y”
 17-14

■ Regression equation of Y on X:
Ye = a + bX
where
Ye : dependent variable
X: independent variable
a & b : constants to be estimated
“It measures the amount of change in Y corresponding
to unit change in X”
 17-15

Method1:Calculating regression equation by

solving two normal equations:-

■ Regression equation of X on Y
X=a+ bY
To determine values of a and b we will solve two normal equations:
∑ X=Na+ b ∑ Y (1)

∑ XY=a ∑ Y+b ∑Y² (2)

■ Regression equation of Y on X
Y=a+ bX
To determine values of a and b we will solve two normal equations:
∑ Y=Na+ b ∑ X (1)

∑ XY=a ∑ X+b ∑X² (2)

 17-16

Method2:-Deviation taken from mean of X

and Y:

■ Regression equation of X on Y:
X - X ¯ = bxy (Y - ¯Y)

where :
X and ¯Y : mean of X and Y
bxy = regression coefficient of X on Y
= ∑xy/ ∑y²
x = (X - ¯X) and y = (Y - ¯Y)
 17-17

Deviation taken from mean of Y on X:

■Regression equation of Y on X:
Y - ¯Y = byx (X - ¯X)

where :
¯X and ¯Y : mean of X and Y
byx = regression coefficient of Y on X

= ∑xy
/ ∑x²
Where x = (X - ¯X) and y = (Y - ¯Y)
 17-18

Method2:-Deviation taken from assumed

mean of X and Y:
■ Regression equation of X on Y:
X - X ¯ = bxy (Y - ¯Y)

where :
¯X and ¯Y : mean of X and Y
bxy = regression coefficient of X on Y
= (N∑dxdy - ∑dx ∑ dy ) /(N ∑dy² - (∑dy)² )
Where dx=X-A and dy =Y-B
■ Regression equation of Y on X:
Y - ¯Y = byx (X - ¯X)

where :
¯X and ¯Y : mean of X and Y
byx = regression coefficient of Y on X
= (N∑dxdy - ∑dx ∑ dy ) /(N ∑dx² - (∑dx)² )
Where dx=X-A and dy =Y-B

*you can also use notations as U=dx and V=dy

 17-19

Method2:-Deviation taken from mean of X

and Y:
■ Regression equation of on Y:
X - X ¯ = bxy (Y - ¯Y)

where :
¯X and ¯Y : mean of X and Y
bxy = regression coefficient of X on Y
= r (σx / σy )

■ Regression equation of Y on X:
Y - ¯Y = byx (X - ¯X)

where :
¯X and ¯Y : mean of X and Y
byx = regression coefficient of Y on X
= r (σy / σx )
 17-20

Case 1: The following data (in Rs Crores)

gives expenditure on advertisement and sales
of a firm:

Advertisement Sales (y)

Expenditure (x)

Mean 10 90

S.D 3 12

Correlation Coefficient =0.8

 17-21

Regression equation in a bi-variate grouped frequency:

 The procedure is same as of deviation method. The modified

formulae are:
bxy = (N∑fdxdy - ∑fdx ∑ fdy )h /( ∑fdy² - (∑fdy)² ) k
byx = (N∑fdxdy - ∑fdx ∑ fdy ) k /( ∑fdx² - (∑fdx)² ) h

*notations can be used in different manner also

By taking U=dx and V=dy
 17-22

Statistics Associated with Bivariate

Regression Analysis
■ Regression coefficient. The estimated b is usually referred to as
the regression coefficient.
■ The coefficient of correlation is the geometric mean of two
regression coefficients.i.e. r² = byx * bxy
■ (byx + bxy )/2 ≥ r
■ If one of the regression coefficient is greater than unity, the other
must be less than unity.(b’coz rcannot be greater than 1)
■ Both regression coefficient will always have same sign.
■ If byx is positive , then bxy will also be positive and so that r.
■ If byx is negative , then bxy will also be negative and so that r.
■ Regression coefficients are independent of change of origin but
not scale.
 17-23

Case2:-
A departmental store gives in service training to its salesman which is
followed by a test .The management is considering whether it
should terminate the service of any salesman who does not do well
in the test .The following data gives the test score and sales by 9
salesman during a certain period:
Test scores: 14 19 24 21 26 22 15 20 19
Sales(’00Rs): 31 36 48 37 50 45 33 41 39
Calculate the coefficient of correlation between the test scores and the
sales. Does it indicate that the termination of service of low test
scores is justified ?If the firm wants a minimum sales volume of
Rs.3000,what should be the minimum test score.what is the
probable sales volume of salesman making a score of 28.
 17-24

Case3:

The following table give age of car of a certain make and

annual maintenance costs. Obtain the regression equation
for cost related to age:
Age of car(in years):2 4 6 8
Maintenance cost :10 20 25 30
(in Rs. hundred)
Also estimate the annual maintenance cost for ten year old
car.

Ans:Rs.3750
 17-25

Case4:
■ For some bivariate data following results were obtained:
Mean value of variable X=53.2 and of Y=39.5.Regression
coefficient of Y on X=-1.5 and regression coeff. of X
on Y=-0.38.What should be the most likely value of X
when Y=50.Also find the value of r.