Regression & Correlation Statistics

Chapter # 14

Relations between Variables:-

The concept of a relation between two variables such as family income and family
expenditure for housing is a familiar one.
Types of Relation between Variables:-
There are two types of relation between variables which are following.
1) Functional Relation
2) Statistical Relation
Functional Relation:-
A functional relation between two variables is a perfect relation, where the value on the
dependent variable is uniquely determined from the value of the independent variable.
A functional relation is expressed by a mathematical formula. If ‘x’ is the independent
variable and ‘y’ is the dependent variable, a functional relation is of the form y=f (x )
Characteristic of Functional Relation:-
The observation, when plotted on a graph, all fall directly on the line or curve of the
functional relationship.
Statistical Relation:-
A statistical relation is a relation where the value of the dependent variable is not uniquely
determined when the level of the independent variable is specified.
A statistical relation, unlike a functional relation, is not exact. The value of ‘y’ is not
uniquely determined from knowledge of ‘x’.
Characteristic of Functional Relation:-
The observation, when plotted on a graph, do not all fall directly on the line or curve of the
functional relationship.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 1

 Regression & Correlation Statistics
Scatter Diagram:-
A scatter diagram is a plot of all the ordered pairs of values of two variables on a
Co – ordinate axis system. The values of one variable are plotted along the horizontal axis
and the values of other variable on vertical axis.
OR
The scatter diagram is a set of points in a rectangular co – ordinate system, where each
point represents an observed pair of values.
OR
A graph of the pairs of observation on two variables is called a scatter diagrams.
OR
A graphic device used to summarize visually the relationship between tow variables.
Note:-
When constructing a scatter diagram, always plot independent variable, if there is
one, on the horizontal axis the dependent variable on the vertical axis. However, if there is
no independent – dependent distinction, either variable can go on the horizontal axis and
the other on vertical axis.
Regression:-
Regression investigates the dependence of one variable the dependent variable on one or
more other variable(s) the independent variables and provides an equation to be used for
estimating and predicting the mean or average value of the dependent variable in terms of
known or fixed values of the independent variable(s).
OR
Regression is the study of dependency of one variable on the other variable.
OR
When we predict the value of dependent variable with the help of one or more independent
variables, it is known as regression.
OR
Regression is a process by which we estimate the value of dependent variable on the basis
of one or more independent variables.
Curve Fitting:-
Curve fitting is a process of estimation the parameter of the population regression function
from an observed sample

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 2

 Regression & Correlation Statistics
Simple Regression:-
Simple regression investigates the dependence of one variable Y on a single independent
variable X and provides an equation to be used for estimating and / or predicting the mean
or average value of the dependent variable in terms of known or fixed values of the
independent variable.
OR
When we study the dependence of one variable on a single independent variable is called
simple regression.
OR
A regression equation that includes one independent variable and one dependent variable
Multiple Regressions:-
When we study the dependence of one variable on two or more then two variables is called
multiple regressions.
Regressand:-
In regression process the dependent variable is called regressand. It is also called as the
response variable or the predictand variable of the dependent variable or the explained
variable.
Regressor:-
In regression process the independent variable is called as the regressor. It is also called as
the predictor variable or the independent variable or the controlled variable or the
explanatory variable.
Regression Analysis:-
The technique used to develop the equation and provide the estimates is called regression
analysis.
Linear Regression:-
Regression analysis involving one independent variable and one dependent variable
in which the relationship between the variables is approximated by a straight line is known
as linear regression.
OR
If the simple regression describes the dependence of the expected value of the
dependent variable ‘y’ as a linear function of the independent random variable ‘x’ then the
regression is called linear regression or simple linear regression.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 3

 Regression & Correlation Statistics
Sample Simple Linear Regression Modal

Let denote the values of independent non – random variable X, and

denote the observed values of the dependent variable Y associated with .

The sample simple linear regression model based on the sample data consisting of “n” pairs

is given by

Where “a” and “b” are the point estimates of the intercept and slop of the population

regression and are the residuals errors

Linear Regression Coefficient Slop of Regression Line:-
The simple linear regression coefficient is the relative change in the expected value of the
dependent random variable with respect to a unit increase in the independent non – random
variable.
OR
The average change in the dependent variable for a unit change in the independent variable
is called regression coefficient.
The regression coefficient may be positive or negative, depending on the relationship
between the two variables.
Non – Linear Regression:-
Non – linear regression is a procedure for fitting data to equations which are non – linear in
the parameters.
Regression Line OR Regression Equation:-
A straight line the best represents the relationship between two variables.
OR
An equation that predicts the value of the dependent variable based on the value of one or
more independent variables is called a regression equation or regression line.
Least Squares Principle:-
The principle of least squares says that the sum of squares of the residuals of observed
values from their corresponding estimated values should be the least possible. This
principle was given by a French mathematician Adrien Legendre.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 4

 Regression & Correlation Statistics
Least Squares Methods:-
The least squares methods is the procedure used to develop the estimated regression
equation.
OR
The method of least squares consists of determine the values of unknown parameters that
will minimize the sum of squares of residuals.

The residuals of the i-th point from the estimated regression line is

Sum of squares of residuals is

Residual:-
The difference between the observed value of the dependent variable and the value
predicted using the estimated regression equation is called residual
Population Regression Function:-
Let Y be the dependent variable and X be the independent variable and

When we look for a relationship is the expected value of Y for a given value of x, the
population regression function is given by

= f(x)
Where f(x) is a function of independent variable x and may be of the linear,
quadratic, exponential or any other form.

We may assume that population regression function is a linear function of

Where

= a pre – determined value of a non – random variable

= associated with “x” a random variable

= the y – intercept of the population regression curve

= the slop of the population regression curve also known as population regression
coefficient

= the deviation ( - ), in the population regression error

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 5

 Regression & Correlation Statistics

Let Let

are are
n observed value of a random variable Y, with n observed value of a random variable X, with
their associated ‘x’ values, then estimated their associated ‘y’ values, then estimated
regression line “Y on X” is regression line “X on Y” is

Where is the estimated value of Where is the estimated value of

‘a’ (y intercept) is the point estimate ‘α’ and ‘a’ (y intercept) is the point estimate ‘α’ and
‘b’ (slop) is the point estimate of ‘β’. The least ‘b’ (slop) is the point estimate of ‘β’. The
squares estimates of coefficients α and β are least squares estimates of coefficients α and β
the values a and b for which sum of squared are the values a and b for which sum of
residuals squared residuals

is minimum. The resulting estimates are is minimum. The resulting estimates are

Least Squares Estimation:-

More Usual Situations

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 6

 Regression & Correlation Statistics

Regression coefficients

Y on X( ) X on Y( )

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 7

Regression & Correlation Statistics

x – intercept( ) x – intercept ( )

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 8

 Regression & Correlation Statistics
Independent variable:-
A variable that provides the basis for estimation is called the predictor or explanatory
variable or independent variable.
OR
The variable that is predicting or explaining the other variable is called regressor variable
or independent variable.
Dependent variable:-
The variable that is being predicted or estimated is known as dependent variable or
regressand variable or predictand variable or response variable.
Regression coefficient β
The regression coefficient β indicates the expected change in dependent variable Y
associated with one – unit increase in the value of the independent variable X.
Properties of the Least Square Line:-
The line fitted by the method of least squares has a number of properties. Some
properties which are following.
(1) The sum of the residuals is zero.

That is.

(2) The sum of the squares of the residuals is minimum

That is
(3) The regression line always passes through the point of means ( ), the center of
gravity of the observed data.

That is wherever , we have

(4) The sum of the observed values equals the sum of the fitted values

That is

i. The mean of the fitted values is the same as the mean of the observed

values

That is

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 9

 Regression & Correlation Statistics
ii. Is also equal to zero.

Correlation:-
Correlation is measure of the degree of linear association between tow variables.
OR
Correlation measures the strength of a relationship between variables.
OR
The degree or strength of relationship (interdependence) between the variables is called
correlation.
Correlation coefficient:-
The numerical value of correlation is called correlation coefficient. It is denoted by ‘r’.
OR
A measure that expresses the extent to which two variables are related.
OR
A descriptive measure of the degree of linear relationship between ‘X’ and ‘Y’ is called
correlation coefficient.
Correlation Analysis:-
A group of techniques to measure the strength of the association between two variables.
Positive Correlation:-
When the values of two variables move in the same direction so that an increase or
decrease in the value of one variable is associated with an increase or decrease in the value
of the other variable, correlation is said to be positive.
OR
If both the variables are moving in same direction (increase or decrease), then it is said to
be positive or direct correlation.
Example:-
Ages and heights of children.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 10

Regression & Correlation Statistics

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 11

 Regression & Correlation Statistics
Negative Correlation:-
When the values of two variables move in the different direction so that with an increase in
the value of one variable the value of the other variable decreases and with a decrease in
the value of one variable the value of other variable increases, correlation is said to be
negative.
OR
If both the variables are moving in opposite direction it is called negative or inverse
correlation.
Example:-
Increase in the Supply of a commodity decrease its price.
No Correlation:-
When two variables have zero correlation, it is called as no correlation.
OR
If the change in one variable does not affect the other variable, then there will be no
correlation.
Example:-
The head sizes and I.Q’s of persons.
Curvilinear Correlation:-
When correlation between two variables represents a curve that is not a straight line, then
the correlation is said to be curvilinear correlation.
Linear Correlation:-
Linear correlation is one where the ratio of variations in the related variables is constant.
Non – Linear Correlation:-
Non – linear correlation is one where the ratio of variations in the related variables is
fluctuation.
Perfect Correlation:-
If the relationship between variables is such that with an increase or decrease in the value
of one, the value of the other increase or decrease in a fixed proportion, correlation between
them is said to be perfect correlation.
Perfect Positive Correlation:-
If both the series move in the same direction and the variations are proportionate there
would be perfect correlation between them, is said to be perfect positive correlation.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 12

 Regression & Correlation Statistics
Perfect Negative Correlation:-
If the two series move in reverse directions and the variations in their values are always
proportionate, it is said to be perfect negative correlation.
Product Moment Co – efficient Of Correlation:-
A numerical measure of strength in the linear relationship between any two variables is
called the Pearson’s product moment correlation co – efficient of simple correlation.
The simple linear correlation for ‘n’ pairs of observations is defined by

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 13

 Regression & Correlation Statistics
Properties of Simple Correlation Coefficient ‘R’:-

The sample correlation coefficient ‘r’ has the following properties.

(1) Correlation coefficient is symmetrical with respect to the variables X and Y

That is

(2) Correlation coefficient is the covariance of values of the two variables X and Y

measured in standard units.

That is

(3) Correlation coefficient is always lies between –1 and + 1

That is

(4) is the geometric mean of the two regression coefficients and

That is

Thus

(5) Correlation coefficient is zero when one of the variables X and Y is constant.

(6) The magnitude of the sample correlation coefficient is independent of change

of origin and scale.

a. The value of ‘r’ remains unchanged if constants are added to or subtracted from the

values of the variables or if the values of the variables are multiplied or divided by

constants having the same sign.

b. But the value of ‘r’ changes in sign only if the values of the variables are multiplied

or divided by constants having opposite signs,

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 14

Regression & Correlation Statistics

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 15

Regression & Correlation Statistics

1) What is regression?
Ans. It’s a technique to investigate the relationship between variables for prediction and
estimation.
2) Define independent variable (Regressor).
Ans. The variable that forms the basis of estimation or prediction is called ind4ependent
variable
Example: - Distance can be predicted from amount of petrol, so petrol is
independent Variable.
Note: - Independent variable is also called explanatory variable, predictor or
nonrandom variable
3) What is dependent variable?
Ans. The variable whose resulting value depends upon the value of the independent variable
is called independent variable.
Example: - distance can be predicted for amount of petrol, so distances is
dependent variable.
Note: - Dependent variable is also called explained variable, predicted or
random variable.
4) Differentiate between functional relationships.
Ans. In functional relationship the value of dependent variable can be uniquely determined
form the value of the independent variable.

Example: - the area of circle can be determined by while in statistical

relationship value of dependent variable can’t be uniquely
determined from independent variable
Example: - A person’s weight can’t be uniquely determined by his height.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 16

Regression & Correlation Statistics
5) What is regression analysis?
Ans. Regression analysis is a method of estimating an average relationship between variables
to estimate the average value of dependent variable from give values of other variables
are called independent variable
OR
Regression analysis provides an equation that can be used to estimate the average value
of dependent variable from given values of independent variable.
6) What is scatter diagram?
Ans. The scatter diagram is a set of points in a rectangular co –ordinate system where each
point represents an observed pair of values.
7) What is regression curve?
Ans. The regression curve is the locus of the expected value of the response variable for
given value of the regressor.
8) What is curve fitting?
Ans. The process of determining the equation or estimating the parameters appearing in the
equation of approximating curve is called curve fitting.
9) Define principle of least squares?
Ans. According to the principle of least squares the best or plausible values of the unknown
are those for which error (residuals) sum of squares is least
10) Define least squares fit?
Ans. Among all curves approximating a given data, the curve is called a least squares fit for
which the sum of squares of the residuals is the least.
11) Write down source of error in regression model.
Ans. The source of error in regression model which are following.
 Specification error
 Sampling error
 Measurement error
12) What are residual?
Ans. The difference between observed and estimated values is called residual.

13) What is simple regression?

Ans. The study of the dependence of a variable on a single independent variable is called
simple linear regression.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 17

Regression & Correlation Statistics

14) Define simple linear regression.

Ans. When the dependence is represented by a straight line equation, the regression is said to
be linear.
OR
When dependent variable is given as a linear function of independent variable, this is
called simple linear regression.
15) Define simple linear regression coefficient.
Ans. The simple linear regression coefficient indicates the relative change in the expected
value of the dependent random variable with respect to a unit change increase in the
independent not random variable
OR
It indicates the change in the mean of the probability distribution of dependent variable
for per unit increase in independent variable.
16) What is covariance?

Ans. It measures the nature of relationship between variables. It is denoted by and given
as

17) Write down the properties of regression coefficient.

Ans. The properties of regression coefficient which are following
 The geometric mean of two regression coefficients is equal to correlation
coefficient.
 Two regression coefficients have same sign.
 Regression coefficient is independent of origin but not scale.
 If one of the regression coefficients is greater than one then other must be less
than one.

18) For two variables X and Y criticize.

Ans. Bothe regression coefficients cannot be greater than one.

19) For two variables X and Y criticize.

Ans. Tow regression coefficients always have same sign.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 18

Regression & Correlation Statistics
20) Describe the properties of the least squares regression line.
Ans. The properties of the least squares regression line which are following.

 The sum of the residuals is zero, i.e.

 The sum of the observed values equals to the sum of fitted values i.e.

 The sum of the squares the residuals is the least.

 The regression line always passes through the

21) Write down assumption of the linear regression model.
Ans. The assumptions of the linear regression model which are following.

 The expected value of error term is zero i.e.

 The variance of error term is constant i.e.

 Error terms are independent of each other i.e.

 Error terms are normally distributed with mean ‘0’ and variance ‘1’
22) Write down simple linear population regression model?
Ans. The simple linear regression population models are following.

Where

= a pre – determined value of a non – random variable

= associated with “x” a random variable

= the y – intercept of the population regression curve

= the slop of the population regression curve also known as population

regression coefficient

= the deviation ( - ), in the population regression error

Taking expectation on both sides

Expected value of Y for given X

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 19

Regression & Correlation Statistics

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 20

Regression & Correlation Statistics
23) Write down the sample regression model.
Ans. The simple linear regression population models are following.

Taking expectation on both sides

24) What is correlation?

Ans. The degree of relationship between variables is called correlation.
OR
The interdependence between variables is called correlation
25) What is linear correlation?
Ans. The degree of linear relationship between variables is called linear correlation.
26) Differentiate between regression and correlation problems, giving examples.
Ans. Regression is a technique to predict and estimate the value of dependent variable
from a give value of independent variable.
Example we can estimate the distance from a given value of petrol.
Correlation is a statistical method to determine whether a relationship between
variable exists.
Example: - Relationship between height and weight.
27) What is correlation coefficient (pearson product moment correlation coefficient)?
Ans. The correlation coefficient is a measures the degree and direction of relationship
between variables in standard units. It is denoted by “r”.
OR
It is measure of the strength of the linear relationship between two variables. The

sample linear correlation coefficient for “n” pairs of observations usually

denoted by the letter “r” is define by

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 21

Regression & Correlation Statistics
Mathematical formula

28) Define positive correlation

Ans. The correlation is said to be positive (or direct)
if the two random variables tend to move in
the same direction i.e. increase (or decrease)
simultaneously.
29) Define perfect positive correlation
Ans. The correlation is said to be perfect positive
if the relationship between the two variables is
perfectly linear with positive slop.
Note: - In this case all the point lie on the line
30) Define negative correlation
Ans. The correlation is said to be negative (or inverse)
if the two random variables tend to move in
the opposite direction
31) Define perfect negative correlation
Ans. The correlation is said to be perfect negative
if the relationship between the two variables is
perfectly linear with negative slope.
Note: - In this case all the point lie on the line.
32) Define zero or no correlation
Ans. There is no correlation between two variables
if there does not exit any relationship between
two variables.
33) Are regression and correlation analysis make assertions about causality?

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 22

Regression & Correlation Statistics
Ans. No, these techniques don’t inform that there is any causes and effect relationship
between variables
34) Write down the properties of correlation coefficient
Ans. For two variable X and Y the properties of correlation coefficient are following

 r is symmetrical with respect to the variables i.e.

 r is the covariance of values of the two variables measured in standard units i.e.

 r always lies between –1and +1.

 The magnitude of r is independent of change of origin and scale.

 r is geometric mean of two regression coefficients i.e.

 r is zero when one of the variable X or Y is constant.
35) Interpret the meaning when “r = -1”, “r = 0”, “r = +1”
Ans. r = -1 There is perfect negative correlation between the variables.
r = 0 There is no correlation between the variables
r = +1 There is perfect positive correlation between the variables
36) Criticize correlation coefficient “r = -1.96”
Ans. Correlation coefficient “r” can never be less than –1
37) Criticize correlation coefficient “r = +2.96”
Ans. Correlation coefficient “r” can never be greater than +1
38) When do the two regression lines coincide?
Ans. When there is perfect linear relationship between variables i. e.
39) When do the two regression lines at right angles?
Ans. When there is no relationship between variables i.e. r = 0.
40) Write down the relationship between correlation coefficient and regression
coefficients
Ans. The relationship between correlation coefficient and regression coefficients are
following

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 23

Regression & Correlation Statistics


41) What is the relationship between correlation coefficient and regression line
Ans. The correlation coefficient ‘r’ is the slop of the regression lines for standard scores.

42) Given then find

Ans.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 24

Regression & Correlation Statistics

43) Given fine .

Ans.

44) Given find

Ans.

45) Given fine

Ans.

46) Given fine

Ans.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 25

Regression & Correlation Statistics
47) Given find
Ans.

48) If find the

Ans.
Note: - Regression coefficient is independent of change of origin

49) For then find

Ans.

50) Given two regression lines find

Ans. As two regression lines always passes through ( )

(i)

(ii)
Multiplying (i) by 5 and subtracting (ii)

Put in equation (i)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 26

Regression & Correlation Statistics
So

51) For a given set of data, find the product moment

correlation coefficient for the data.
Ans. We’re given

52) For a given set of data, we have find

Ans.

53) Given find

Ans.

54) Given find

Ans.

Squaring on both sides

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 27

Regression & Correlation Statistics

55)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 28

Regression & Correlation Statistics
56) The least squares regression lines are find the
product moment correlating coefficient for the data.
Ans. As we know that

57) Given find two regression coefficients.

Ans.

58) Given find the regression equation of

“Y on X”
Ans.

59) Given find the regression equation of

“X on Y”
Ans.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 29

Regression & Correlation Statistics
60) What is the method of least square?
Ans. The method of least squares consists of determining the values of unknown parameter
that will minimize the sum of squares of residuals.
61) what is ment by curve fitting
Ans. The process of determining the equation or estimating the parameters appearing in the
equation of approximating curve is called curve fitting.
62) Mr. Ali used linear regression to help him understand his monthly telephone bill.

The line of best fit was “x” is the number of long – distance calls
made during a month and “y” is the total telephone cost for a month. In terms of
number of long distance calls and cost, explain the meaning of the y – intercept
1420 and the slope 76.8.
Ans. When no long –distance are made there is still the monthly phone charge of Rs. 1420
and Rs. 76.8 is the rate at which the total phone bill will increase for each additional
long distance call.
63) What does the value or “r” indicates?
Ans. The value of the sample linear correlation coefficient indicates the degree and direction
of linear relationship between variables.
64) Describe the significance of the algebraic sign and the magnitude of “r”
Ans.
1.0 
 The linear correlation coefficient ‘r’ may assume 
 Strong
values from “–1” through zero to “+1”. 0.8 

 When ‘r’ is close to zero, we conclude that there 0.8 

is little or no correlation between the variables.  Moderate
0.5 

 Positive values of ‘r’ suggest that the two
0.5 
variables are positively linearly correlated 


 Negative values of ‘r’ suggest that the two variables 0  Week

are negatively linearly correlated 
 0.5 

 As the calculated value of ‘r’ changes from ‘0’ toward
 0.5 
either “+1.0” or “-1.0”, it indicates an increasingly 
 Moderate
stronger linear correlation between the two variables.  0.8
 Following figure shows a useful way to describe the  0.8 

strength of linear relationship based on “r”  Strong
 1.0 


Raja Wajid Ali (Subject Specialist) call: 0322-4461550 30

Regression & Correlation Statistics
65) Between what limits must the product moment coefficient of correlation lies?
Ans. The product moment coefficient of correlation lies between “-1 ant +1”.
That is – 1 ≤ r ≤+1

66) Why do the regression coefficients and the linear correlation coefficients ‘r’
have the same sign?

Ans. The sign of is the same as that of any regression coefficient

the reason is that the numerators of their computational formulas

are the same and determine the sign of and ‘r’ and the denominators are always
positive.
67) Is the following statement correct? Explain why or why not. A correlation
coefficient of zero implies that there is no relationship exists between the two
variables understudy.
Ans. The value of ‘r’ is a measure of extent to which X and Y are linearly related. A value of
‘r’ close to zero does not rule out any strong relationship between X and Y; there could
still be a strong relationship but one that is not linear. Therefore, we should not
conclude that there is no relationship whatsoever just because ‘r’ is near to zero.
68) Difference between correlation and regression?
Ans.
 In correlation analysis we use both variables as random.
 In regression analysis we take one variable as fixed and the other as random.
 Correlation coefficient measures degree of linear relationship between two variables.
 Whereas the objective of regression is to study the “nature of relationship” between the
variables.
 so that we will be able to predict the values of one variable on basis of on another
69) What is the purpose of simple linear correlation
Ans. The purpose of simple linear correlation is to determine whether or not two or not tow
variables are related.
70) What are the simple linear correlation measures?
Ans. The simple linear correlation measures the strength or closeness of linear relationships
between two variables.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 31

Regression & Correlation Statistics

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 32

Regression & Correlation Statistics
71) Who gave the least squares principle?
Ans. This principle was given by a French mathematician Adrien Legendre.
72)
Ans.
73)
Ans.
74)
Ans.
75)
Ans.

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 33

Regression & Correlation Statistics

Each question has possible four answers. Choose the correct answer and encircle it.
1) A _____ relation between two variables is a relation where the value of the dependent
variable I uniquely determined from the value of the independent variables
a) Fractional b) Functional c) Cubic d) Squares
2) The study of relationship between variables for prediction and estimation
a) Regression b) Time Series c) Correlation d) All
3) A functional relation is expressed by a formula
a) Mathematical b) Chemical c) Biological d) Empirical
4) The variable that forms the basis of estimation or prediction
a) Regressand b) Regressor c) Dependent Variable d) Explained Variable
5) If x is an independent variable and y is a dependent variable then the functional
relation is of the form

a) b) c) d)
6) If y is an independent variable and x is a dependent variable then the functional
relation is of the form

a) b) c) d)
7) If ‘t’ is an independent variable and ‘s’ is a dependent variable then the functional
relation is of the form

a) b) c) d)
8) If ‘s’ is an independent variable and ‘t’ is a dependent variable then the functional
relation is of the form

a) b) c) d)
9) For a given value of ‘x’, the functional relation ______ gives the corresponding value
of ‘y’

a) b) c) d)
10) The variable that forms the basis of estimation or prediction is called the
a) Regressor b) Regressand c) Regression Curve d) None of these
11) The variable whose resulting value depends upon the selected value of the
independent variable is called the

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 34

Regression & Correlation Statistics
a) Regressor b) Regressand c) Regression Curve d) None of these
12) The locus of the expected value of the response variable for given values of the
regressor variable is called
a) Regressor b) Regressand c) Regression Curve d) None of these
13) Regressor variable is also called
a) Predictor b) Dependent c) Uncontrolled d) None of these
14) Regressor variable is also called
a) Dependent b) Independent c) Uncontrolled d) None of these
15) Regressor variable is also called
a) Dependent b) Controlled c) Uncontrolled d) None of these
16) Regressor variable is also called
a) Dependent b) Uncontrolled c) Explanatory d) None of these
17) Regressand variable is also called
a) Dependent b) Independent c) Explanatory d) Regressor
18) Regressor variable is also called
a) Independent b) Controlled c) Explanatory d) All
19) All values for directly on the line
a) Functional Relation b) Statistical Relation
c) Mathematical Relation d) ‘a’ and ‘c’
20) The dependent variable is not uniquely determined
a) Functional Relation b) Statistical Relation
c) Mathematical Relation d) Deterministic Relation
21) The observations don’t fall directly on the line or curve
a) Statistical Relation b) Regression Relation
c) Probabilistic Relation d) All
22) The relationship between yield and quantity of fertilizer is
a) Functional b) Statistical c) Deterministic d) All
23) The height of a son related to the height of his father
a) Probabilistic b) Statistical c) Regression d) All

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 35

Regression & Correlation Statistics
24) Which is true for a mathematical model?
a) Contains the random error b) Dos not contain the random error
c) Contains residuals d) With error term
25) A statistical model
a) Contains the random error b) Dos not contain the random error
c) Show as exact relationship d) None of these
26) The method of estimating an average relationship between variables
a) Regression b) Time Series c) Correlation d) Computer
27) The best fitting curve has sum of squares of residuals
a) Minimum b) Maximum c) Largest d) None of these

28) The errors are often called

a) Residuals b) Deviations c) Prediction Errors d) All
29) Error arising from the omission of one or more relevant independent variables
a) Measurement Error b) Sampling Error c) Specification Error d) Random Error
30) A simple linear regression model contains
a) One Variable b) Two Variables c) More than two variable d) (a) & (b)
31) One regression line is horizontal and the other regression line is vertical.
a) r = 0 b) r < 0 c) r > 0 d) r=1
32) Which is a parameter of linear regression model?

a) α, β b) c) d) All
33) The least squares regression line minimizes
a) Calculation mistakes b) SSE c) Spelling mistakes d) None of these
34) The change in the dependent variable Y corresponding to a unit change in the
independent variable ‘x’ is measured by

a) b) c) d) None of these
35) Regression coefficient is independent of
a) Origin b) Scale c) Both ‘a’ & ‘b’ d) Unit of Measurement
36) Two regression coefficient always have
a) Opposite Signs b) Same Signs c) No Sign d) None the these

37) If is greater than one, is also

a) < 1 b) also >1 c) = 1 d) None of these

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 36

Regression & Correlation Statistics
38) If then would be
a) – 1.34 b) 0.50 c) 1.34 d) - 1.00

39) The sign of has always

a) Same sign b) Opposite sign c) No sign d) + , +

40) For is always equal to

a) Zero b) Greater than zero c) More than zero d) Undefined

41) For

a) b) c) d)
42) The regression line between X and Y always passes through

a) b) origin c) d) none of these

43) It measure the linear mutual variability of the two variables
a) Covariance b) Skewness c) Dispersion d) Central Tendency
44) If the variables tend to move in same direction than sign of covariance is
a) Negative b) Positive c) No Sign d) None of these
45) If the variables tend to move opposite direction than sig of covariance is
a) Negative b) Positive c) No Sign d) None of these
46) Null Y are independent variables than Cov(X,Y) =
a) Negative b) Positive c) Zero d) None – Zero

47) If then
a) -0.84 b) 0.84 c) 0.42 d) None of these

48) If , “u=-2x” and “v=4y” then

a) -0.84 b) 0.84 c) 0.42 d) Undefined
49) The value of the correlation coefficient is always in the range.
a) 0 to 1 b) -1 to +1 c) -1 to 0 d) None of these
50) Two regression lines are perpendicular to each other when

a) b) c) d)
51) The regression lines coincide if

a) b) c) d)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 37

Regression & Correlation Statistics
52) If two regression lines are & then the correlation coefficient
between ‘x’ and ‘y’

a) b) c) d)
53) If two regression coefficients are ‘0.8 & ‘0.2’ then value of coefficient of correlation is
a) 0.16 b) -0.16 c) 0.40 d) -0.40
54) The correlation coefficient ‘r’ is the _______ mean of the two regression coefficients
a) Geometric b) Arithmetic c) Harmonic d) None of These
55) The correlation coefficient of two variables would be
a) -1.95 b) 1.95 c) 0.95 d) None of these
56) ‘r = 0’ indicates that the two variables are linearly
a) Independent b) Dependent c) Related d) None of These

57) The negative sign of indicate that X and Y

a) Directly Related b) Inversely Related
c) Move in Same Direction d) No Relationship

58) If then and are

a) Negative b) Positive c) Zero d) None of These
59) The geometric mean of two regression coefficients
a) Correlation Coefficient b) Regression Coefficient
c) Coefficient Of Skewness d) None Of These
60) If one of the variables is constant the correlation coefficient is
a) Less Than One b) More Than One c) Equal to One d) Zero
61) Covariance of the two variables measured in standard units
a) Correlation Coefficient b) Regression Coefficient
c) Coefficient Of Skewness d) None Of These
62) The magnitude of correlation coefficient doesn’t change if
a) Origin Is Changed b) Scale Is Changed
c) Both ‘a’ & ‘b’ d) None of these

63) The positive sign of indicate that X and Y

a) Inversely b) Directly c) Positively d) Unexplained
64) A negative correlation indicates that one variable increases as other
a) Decreases b) Increases c) Unchanged d) None of These

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 38

Regression & Correlation Statistics
65) A positive correlation indicates that one variable increases as other
a) Decreases b) Increases c) Unchanged d) None of these
66) In correlation analysis two regression lines always intersect at

a) Y – Intercept b) Origin c) ( ) d) None of these

67) The slop of the regression line also known as
a) Regression Coefficient b) Y – Intercept
c) Ordinates d) None of these
68) If the two regression coefficients are negative, then correlation coefficient is
a) Positive b) Negative c) No Sign d) Does Not Exist
69) If the two regression coefficients are positive, then correlation coefficient is
a) Positive b) Negative c) No Sign d) Does Not Exist
70) The two regression lines are at right angle when the correlation coefficient is
a) Zero b) Positive c) Negative d) None of these
71) r = 1 indicates
a) Perfect positive correlation b) Perfect negative correlation
c) Nonsense correlation d) No correlation
72) the correlation coefficient “r” is
a) Independent of unit of measurement b) Independent of origin
c) Not independent of scale d) ‘a’ & ‘b’
73)
a) b) c) d)
74)
a) b) c) d)
75)
a) b) c) d)
76)
a) b) c) d)
77)
a) b) c) d)
78)
a) b) c) d)
79)
a) b) c) d)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 39

Regression & Correlation Statistics
80)
a) b) c) d)
81)
a) b) c) d)
82)
a) b) c) d)
83)
a) b) c) d)
84)
a) b) c) d)
85)
a) b) c) d)
86)
a) b) c) d)
87)
a) b) c) d)
88)
a) b) c) d)
89)
a) b) c) d)
90)
a) b) c) d)
91)
a) b) c) d)
92)
a) b) c) d)
93)
a) b) c) d)
94)
a) b) c) d)
95)
a) b) c) d)
96)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 40

Regression & Correlation Statistics
a) b) c) d)
97)
a) b) c) d)
98)
a) b) c) d)
99)
a) b) c) d)
100)
a) b) c) d)
101)
a) b) c) d)
102)
a) b) c) d)
103)
a) b) c) d)
104)
a) b) c) d)
105)
a) b) c) d)
106)
a) b) c) d)
107)
a) b) c) d)
108)
a) b) c) d)
109)
a) b) c) d)
110)
a) b) c) d)
111)
a) b) c) d)
112)
a) b) c) d)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 41

Regression & Correlation Statistics
113)
a) b) c) d)
114)
a) b) c) d)
115)
a) b) c) d)
116)
a) b) c) d)
117)
a) b) c) d)

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 42

Regression & Correlation Statistics

Raja Wajid Ali (Subject Specialist) call: 0322-4461550 43