Master of Science II (Statistics)Examination: Oct/Nov 2016
Semester III (New CBCS)
SLR No. Day
Date Time Subject Name Paper
No. Seat No.
SLR SB
704
Friday
25/11/2016
02.30 PM
to
05.00 PM
Regression Analysis
C
XV
Instructions: Answer any five questions.
Q. No. and Q. No. are compulsory.
Attempt any three from Q. No. to Q. No.
Figures to the right indicate full marks.
Total Marks:70
Q.1 A. Select the correct alternative: 05
Forward selection procedure begins with assumption that there

No regressors in the model All regressors in the
model
Some regressors in the
model
None of these
Multicollinearity in linear regression is concerned with
The error terms The regressors
The response variable
values
The coefficients
In the usual notations, variance of ith PRESS residual is
The largest condition index of is given as
In classical linear regression, the distribution of response variable is

Poisson normal
Binomial Uniform
B. Fill in the blanks: 05
A regression model that is not linear in unknown parameters is called is
called
An observation with large residual is called
In usual notations, var
When distribution of Y is poisson, variance stabilization
transformation is used.
The standardized residual have approximately variance.
Page 1 of 2
C. State whether following statements are true or false: 04
Nonparametric regression requires larger sample size than parametric
regression model.
Residual plots are useful for the detection of auto correlation.
In regression model errors are uncorrelated.
Normal probability plot is used to check normality assumption.
Q.2 A. Define: 06
Studentized residual
Standaralized residual
PRESS residual
B. Write short note on the following 08
Variance stabilizing transformation
Box-Cox power transformation
Q.3 A. State multiple linear regression model. Derive LSE of vector of regression
parameters.
07
B. In usual notations, outline the procedure of testing a general linear
hypothesis.
07
Q.4 A. What are the uses of residual plots? Discuss normal probability plot and plot
of residuals against fitted values.
07
B. Describe orthogonal polynomial to fit the polynomial model in one variable. 07
Q.5 A. Describe variable selection problem. Discuss backward elimination method
for subset selection in regression model.
07
B. What is the need of robust regression? Explain M-estimator and Huber loss
function.
07
Q.6 A. Explain the problem of auto correlation. Describe Durbin-Watson test for
detecting auto correlation. What are its limitations?
07
B. In multiple linear regression model, show that 07
Q.7 A. Explain the problem of multicollinearity in connection with linear regression
models. What are the consequences of multicollinearity on least squares
estimates?
07
B. Describe polynomial models in one variable and two variables. 07
Page 2 of 2