LOYOLA COLLEGE (AUTONOMOUS), CHENNAI – 600 034

BA 22

M.Sc. DEGREE EXAMINATION – STATISTICS

FIRST SEMESTER – November 2008

    ST 1811 – APPLIED REGRESSION ANALYSIS

 

 

 

Date : 11-11-08                 Dept. No.                                        Max. : 100 Marks

Time : 1:00 – 4:00

Section A

Answer All the Questions                                                          (10 x 2 = 20 Marks)

 

1. What are the distributions of the components (y_i , ) of a simple regression model?
2. What do you mean by Hetroscedasticity Property?
3. Explain the Linear Probability model
4. What is the Linear Predictor in a Generalized Linear Model?
5. What is the Identity Link in a Generalized Linear Model?
6. Explain the Simple Regression Model.
7. List the four methods of scaling residuals.
8. Give two examples for nominal variables.
9. Give an example of a variable that is classified as nominal, ordinal and interval variable.

10.     Give an example for Poisson Log linear Model.

 

Section B

Answer Any Five Questions                                                          (5 x 8 = 40 Marks)

 

• Show that the least square estimates ,  of a simple regression

model are unbiased.

• Explain the procedures for finding the confidence interval forand  of a simple regression model.
• Discuss multicollinearity with an example.
• Explain the purpose of Unit Normal Scaling
• Explain Binomial logit model for the binary data
• What are the properties of the least square estimates of the fitted regression model?
• Discuss any two methods of scaling residuals
• Explain the Logistic regression model with an example

 

Section C

Answer Any Two Questions                                      (2 x 20 = 40 Marks)

 

19 (a) Derive V() of a simple regression model.

(b) Write down the test procedures to test the intercept of a simple

regression model.

 

20 (a) Write down the test Procedures to test H₀:  =0 against H₁:  ¹ 0

using Analysis of Variance.

(b) Derive the interval estimation of the mean response of a simple

regression model.

 

21 (a) Fit a Logistic regression model

(b) How do you interpret the Poisson Loglinear model for a count data

 

22 (a) Estimate the parameters of a multiple linear regression model by the

methods Maximum Likelihood Estimation.

(b)  Write short notes on Relative Risk, Odds Ratio and cross product ratio.

 

 

Go To Main page

Latest Govt Job & Exam Updates:

View Full List ...