Homework-Chapter #4 OLS Regression and Inference (Students must show all work/ca
ID: 1110108 • Letter: H
Question
Homework-Chapter #4 OLS Regression and Inference (Students must show all work/calculations to get full credit) 1. Use the GPA1 data. Estimate the model using STATA colGPA = 0 + 1 * hsGPA + 2*ACT + 3*fathcoll + 4"mothcoll + 5*siblings + u and report the results using the usual format. (10 points) a. b. Test the hypothesis that Ho: 1-0 against the Ha:1#0.), briefly explain results of c. Test the hypothesis that Hof!=l against the Ha: briefly explain results of d. Test the hypothesis that Ho:3-4 against the Haf334. briefly explain results of e. Estimate the model using STATA colGPA = 0 + 14sGPA + u and report the f Conduct the F test (using both the test statistic and R-squared approach) and test the g. If there is heteroscedasticity what can you say about the t-test and f-test? Are the tests hypothesis test. (15 points) hypothesis test. (15 points) hypothesis test (15 points) results using the usual format. (10 points) hypothesis that Ho:p2-p3-4-0 against the alternative that at least one is not equal to zero, briefly explain results of hypothesis test. (25 points) valid, briefly explain? (10 points)Explanation / Answer
USE the following command in STATA
a) reg colGPA hsGPA ACT fathcoll mothcoll siblings
b) Given the results from part a) check the p-value for the variable hsGPA
If the associated p-value is very small that is close to 0. The reject the null hypothesis which implies that hsGPA has no effect on colGPA.
c) Now you need to calculate the t- value
t =[ 1(hat)-1]/(SE(1(hat))) ,if the t calculated is sufficiently large then reject the null hypothesis else do not.
d) Now calculate t = (3(hat)- 4(hat))/SE(3(hat)- 4(hat))
If the t calculated is sufficiently large then reject the null else do not reject the null.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.