This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: UBC, ECON 326  033 Solution to 2010 Midterm Examination Question 1 (a) True. ˆ U i ’s are the fitted residuals from a regression with an intercept, and ∑ i ˆ U i = is the normal equation obtained from the OLS firstorder conditions for the intercept. (b) False (in general). ˜ U i ’s are the fitted residuals from a regression without an inter cept, and therefore the OLS firstorder condition corresponding to the intercept does not have to hold. (c) False (in general). EU i = 0 , however, a sample average of (finitely many) U i ’s does not have to be zero. (d) True. By the law of iterated expectation (LIE), E U i X 4 i = E E U i X 4 i  X i = E X 4 i E ( U i  X i ) = E X 4 i · = 0 . (e) False. Since ˜ β 1 is a biased estimator, the GaussMarkov Theorem does not apply in this case. In fact, one can show that ˜ β 1 has a smaller conditional variance given X 1 ,...,X n . Question 2 (a) Since n = 29 and the regression has an intercept, the number of df is 27....
View
Full
Document
This note was uploaded on 03/24/2012 for the course ECON 326 taught by Professor Whisler during the Spring '10 term at UBC.
 Spring '10
 whisler

Click to edit the document details