hw3fall11

# hw3fall11 - (c Test whether there is a linear relationship...

This preview shows page 1. Sign up to view the full content.

STAT5044: Regression and ANOVA Homework #3 Problem# 1. (Sales growth data) A marketing researcher studies annual sales of a product that had been introducted 10 years ago. The data are as follows, where X is the year (coded) and Y is sales in thousands of units: i 1 2 3 4 5 6 7 8 9 10 x i 0 1 2 3 4 5 6 7 8 9 y i 98 135 162 178 221 232 283 300 374 395 Does a linear relation appear adequate here? Do you also believe that all the assump- tions about a simple linear normal regression model are correct for this data? If not, what is the problem ? Justify your claim. Problem# 2. Without using R commends such as “lm”, “bptest”, and “conf”, you program for each problem and solve the problem. Refer to Airfreight breakage (Prob- lem 1 in hw2). (a) Fit the simple linear regression and estimate parameters. (b) Obtain conﬁdence interval under normality assumption of error
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: (c) Test whether there is a linear relationship between x and y. Problem# 3. Refer to Airfreight breakage (Problem 1 in hw2). Without normality assumption, create the distribution of β 1 using bootstrap method and then obtain a 95% conﬁdence interval of β 1 . Compare your conﬁdence interval with the one which obtained under normality assumption. Why diﬀerent? or why similar? NOTE: use the following R commend, “sample(x,n,replace=TRUE)” and “quan-tile(x,probs=c(0.025,0.975)”) Problem# 4. Refer to Sales growth data (Problem 1) Check all assumptions. Perform Brown-Forsy test. Consider 5 or 6 diﬀerent two groups and calculate how many times you obtain constant variance. Compare this result with Brunsch-Pagan test 1...
View Full Document

## This note was uploaded on 01/02/2012 for the course STAT 5044` taught by Professor Staff during the Fall '11 term at Virginia Tech.

Ask a homework question - tutors are online