This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Question 1: Use data “WAGE1.XLS” to estimate the equation: log(wage) = β + β 1 educ + β 2 exper + β 3 tenure 1.Report the results in usual form, i.e. replace β with actual estimates and report the standard error in a bracket below β . 2.What is the interpretation of the slopes: β 1, β 2 β 3 ? 3.State and test weather the return to experience, controlling for education and tenure, is zero against the alternative that it is positive. 4.State and test weather the return to education is 4 folds to the return to tenure. Answer: 1.Report the results in usual form, i.e. replace β with actual estimates and report the standard error in a bracket below β . log(wage) = 0.284 + 0.092 educ + 0.004 exper + 0.022 tenure (0.104) (0.007) (0.002) (0.003) 2.What is the interpretation of the slopes: β 1, β 2 β 3 ? β 1 measures when education increases one year, how many percentage wage would increase. In this case, when education increases by one year, wage would increase 9.2%. The other paramet ers can be interpreted similarly. 3.State and test weather the return to experience, controlling for education and tenure, is zero against the alternative that it is positive. H : β 2 = 0 vs. H 1 : β 2 > 0 This is a onesided test. Under 5% significance lev el, with n=526, the critical value is 1.645. The tstatistic = 0.00412/0.0017 = 2.42. Since the t statistic is bigger than the critical value, we reject the null and conclude that the return to experience, controlling for education and tenure, is positive. 4.State and test weather the return to education is 4 folds to the return to tenure. H : β 1 = 4 β 3 vs. H 1 : β 1 ≠ 4 β 3 In Eviews, we select “view”  “coefficient test”  “wald coefficient test”. In the Wald test window, input “C(2) = 4*C(4)”. Be cautious when you input the correct coefficients. The test results hav e a pvalue of 0.803, which implies that there is a 80% probability that we would observe the Fstatistic if the null is true. Thus we fail to reject the null. Question 2: Newhouses.xls is a quarterly data from 1992Q1 to 2004Q4. It has the following variables: NHS: number of houses sold. IR: mortgage interest rate. DPIPC: disposable personal income per capita. GDP: gross domestic product. 1.Estimate the equation: NHS = β + β 1 IR + β 2 DPIPC 2.Report the results in usual form, i.e. replace β with actual estimates and report the standard error in a bracket below β . 3.What is the interpretation of the slopes: β 1, β 2 ? Answer: 2.Report the results in usual form, i.e. replace β with actual estimates and report the standard error in a bracket below β . NHS = 369.732191.907 IR + 0.179 DPIPC (536.374) (31.697) (0.014) 3.What is the interpretation of the slopes: β 1, β 2 ?...
View
Full
Document
This note was uploaded on 04/16/2008 for the course ECN 3620 taught by Professor Gai during the Spring '08 term at Babson College.
 Spring '08
 gai
 Econometrics

Click to edit the document details