Download Binary Response Models - Introduction to Econometrics - Exam and more Exams Econometrics and Mathematical Economics in PDF only on Docsity!
Ollscoil na hÉireann, Gaillimh GX_____
National University of Ireland, Galway
Repeat Examinations 200 9 /20 10
Exam Code(s) 1EP1, 2EP1, 1MEE1, 1MEM1, 1EE1, 2EE2, 1MPC1, 1MH Exam(s) M.A. & M.Econ.Sc. (Economic Policy Evaluation & Planning), M.A. & Econ.Sc. (Economic & Environmental Modelling), M.Sc. in Health Economics, M.Phil. Module Code(s) EC Module(s) Econometrics Paper No. 1 Repeat Paper External Examiner(s) Professor Robert Wright Internal Examiner(s) Professor John McHale Dr. Mary Silles Instructions: Instructions: 7 Questions, attempt all questions in part A, two questions from part B, and all questions in part C. Duration 3 hours No. of Pages 5 in total Department(s) Economics Course Co-ordinator(s) M. Silles Requirements : MCQ Handout Statistical Tables Graph Paper Log Graph Paper Other Material
Instructions: 7 Questions, attempt all questions in part A, two questions from part B, and all questions in part C. Part A. Answer all questions. Question 1. Briefly state with reason whether the following statements are true, false or uncertain. Be precise. (i) A simple regression analysis of the following model would uncover the ceteris paribus effect of student-teacher ratio on test scores. (2 marks) (ii) From the following regression results a 1% increase in family income is associated with an increase in test score of 0.28 points. (2 marks) (iii) A small p-value is evidence that the null hypothesis is false. (2 marks) (iv) The overall regression F-statistic tests the joint hypothesis that all the slope coefficients are zero in the following model. (v) From the following regression results it is apparent that women achieve a higher rate of return to training than men on average. (2 marks) (vi) The estimated regression function suggests that family income has a diminishing effect on test score results. (2 marks) (vii) It is difficult to find instrumental variables which satisfy the two conditions for a valid instrument (2 marks). (viii) OLS is still BLUE is the presence of imperfect multicolinearity. (2 marks) (ix) Five categorical educational levels are introduced into a regression model using four dummy variables. (2 marks) (x) The following model is estimated by probit and suggests that the probability of passing a test increases linearly with hours of study. (2 marks)
Part C. Answer all questions.
- This question relies on information contained in the table of OLS regressions presented below. The authors are trying to assess the wage differential associated with using computers at work. The data is from the German Qualifications and Career Survey and the U.S. Current Population Survey. Use an appropriate test statistic in your answers where relevant. (40 marks). (i) How would you interpret the coefficient on computer in column (1)? Is the coefficient statistically significant at all conventional levels of significance? ( marks) (ii) In moving from column (1) to column (2) is there a statistically significant change in the coefficient on the variable computer? (8 marks) (iii) Interpret the R-squared presented in column (1) and explain its general usefulness in regression analysis. (8 marks) (iv) Do the results in the above table provide strong evidence that individuals who use computers at work receive higher wages as a result of their computer skills? (8 marks) (v) Suggest at least two alternative empirical strategies which might be used to probe whether the computer pay differential is a genuine consequence of computer use or is a selection effect. (8 Marks)
