hw7 - Problem Set 7 Due Friday Econ 321 Applied...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Problem Set 7 - Due Friday 12/05/08 Econ 321 - Applied Econometrics Fall 2008 Cornell University Prof. Molinari This exam consists of THREE parts totaling 100 points. This is a CLOSED BOOK exam. However, useful formulas are provided on pages 5+. Tables for the Normal, t , and F distributions are provided at the end of the text. During the exam calculators are permitted ± but graphic calculators are prohibited. You will have 150 minutes to complete the exam. SHOW ALL your work as partial credit will be given. Answers without explanations might NOT be given (full) credit even if they are correct. THINK before you answer, and GOOD LUCK! Part 1 25 points (5 points each) In WORDS , brie²y (1 paragraph maximum) describe the following: 1. R 2 : 2. Explained Sum of Squares. 3. Gauss-Markov Theorem. 4. Interpreting the slope coe¢ cient of a log-log regression. 5. F ±Test. 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Part 2 40 points (Each sub-question has the number of corresponding points in brackets) Professors tend to believe that students that regularly skip lectures tend to have lower grade point averages ( GPAs ) compared to students that do not skip lectures. Consider the following regression model for college GPA : GPA = & 0 + 1 hsGPA + 2 ACT + 3 skipped + u where GPA is the college GPA of a student, hsGPA is the high school GPA of a student, ACT is the ACT test score and skipped is the average number of lectures missed per week of a student. In an Ivy League University, professors are divided between those who are particularly convinced that students that regularly skip lectures tend to have lower GPAs; and those who do not believe this conjecture. In order to resolve this dispute, they decide to empirically study the relationship. Using a random sample of 540 students, they obtain the following OLS estimates (standard errors are in parentheses): G b PA i = 1 : 45 + 0 : 395 hsGPA i + 0 : 018 ACT & 0 : 093 skipped R 2 = 0 : 237 (0 : 35) (0 : 085) (0 : 0095) (0 : 054) SST = 19 : 406 1. What percentage of the variation in GPA is explained by high school GPA; ACT test score and average number of lectures missed per week? (5 points) 2. Formally test whether there is any statistical evidence to suggest that GPA depends on the In what way does your test depend on whether you could assume MLR.6? (5 points) to suggest that college GPA depends on ACT level. (5 points) 4. On average, what is the estimated di/erence in GPA for a student who had a high school GPA of 3 : 5 compared to a student who had a high school GPA of 4, holding ACT and the average number of lectures missed per week constant? Is this e/ect big or small? What is the 2
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/18/2009 for the course ECON 321 taught by Professor Molinari during the Fall '08 term at Cornell.

Page1 / 11

hw7 - Problem Set 7 Due Friday Econ 321 Applied...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online