final_practice_problems_v1

final_practice_problems_v1 - 1 Econ 166 Professor Altonji...

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

View Full Document Right Arrow Icon
1 1 Econ 166 Professor Altonji Practice Questions for Final Exam, version 1 , Unless otherwise stated statistical tests are to be performed at the 5% significance level. Problem 1. [15 points total] (1.1) [5 points] True or false and explain. In the regression model Y t = B 1 + B 2 X t +u t , if the assumptions E(u t )=0 and E(u t | X 1 ,X 2 ,…,X N )=0 hold for all t and Cov(u t ,u j )=( ρ) | t-j| , for all t j, ρ > 0, then the OLS estimator for B 2 is unbiased. (1.2) [5 points] True or false and explain (two sentences). Heteroscedasticity means that in a regression model (either simple or multivariate regression) the variance of the error term u i is small. (1.3) [5 points] True or false, explain in two sentences. In order to obtain unbiased OLS estimates in a multivariate regression model the researcher has to make sure that she is using explanatory variables which are uncorrelated to each other. Problem 2 [25 points total] Consider the following regression model for earnings on a sample of 20 individuals in a given occupation. You observe for each individual earnings (Y) in $, years of schooling (S) and IQ-test
Background image of page 1

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

View Full DocumentRight Arrow Icon
2 2 scores (this is a test for “intelligence” where points scored typically range, in random samples, from a min of approximately 50 to a max of approximately 250). Consider the regression model (XX) Y i = B 1 + B 2 S i + B 3 IQ i +u i for i=1,2. ....,20 Different versions of the model have been estimated by OLS providing the following Sample Regression Functions. The dependent variable is Y. In the table the columns contain estimated coefficients and corresponding standard errors (in parentheses) . When the coefficient is missing the corresponding variable has not been used in the regression. constant Schooling S IQ-test IQ R-squared Version A -6418 (3349) 1985.7 (248) ---- 0.78 Version B -29586 (4155) 1640 (151) 268 (43) 0.93 Version C -25488 (11362) --- 438 (109) 0.47 (3.1) [10 points] Consider version B. 3.1.a Explain in one or two sentences the meaning of the coefficient on IQ. 3.1.b Predict earnings for an individual with 12 years of schooling and an IQ-test score of 100.
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 07/03/2008 for the course ECON 166 taught by Professor Josephaltonji during the Spring '07 term at Yale.

Page1 / 8

final_practice_problems_v1 - 1 Econ 166 Professor Altonji...

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