This preview shows pages 1–3. 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: ECON 4261: Introduction to Econometrics Fall 2009 Problem Set 2 Answer Key Exercise 1 Suppose you model a variable Y depending on a nonstochastic variable X , according to the relation Y i = X i + u i . The errors are iid with mean zero and variance 2 . Considering estimating by the slope of a line between the origin and one out of the n plotted observations ( X i ,Y i ) . (a) Is this estimator unbiased? Yes. Note that in our case b = slope between (0 , 0) and ( X i ,Y i ) = Y i X i . Then E ( b ) = E Y i X i = E X i + u i X i = E X i X i + E u i X i = E ( )  {z } + 1 X i E ( u i )  {z } =0 = . (b) Find the variance of this estimator. Again, by construction var ( b ) = var Y i X i = var X i + u i X i = var + u i X i = var u i X i = 1 X 2 i var ( u i ) = 2 X 2 i . 1 (c) Which of the observations would you choose to calculate the estimate? Justify your answer. Looking at the answer in (b), we would pick the observation with the greatest X i , since this would minimize the variance of the estimator. Exercise 2 You are given 30 pairs of observations ( X i ,Y i ) which are to be represented by the following model Y i = + 1 X i + u i , and the errors are iid with mean zero and variance 2 . Suppose you know that R 2 = 0 . 25, n i =1 ( X i X ) 2 = 49 and n i =1 ( Y i Y ) 2 = 49. Calculate the OLS estimate for 1 (assume is positive). From class we got the following formula: R 2 = ( b 1 ) 2 n i =1 ( X i X ) 2 n i =1 ( Y i Y ) 2 . 25 = ( b 1 ) 2 (49) (49) . 25 = ( b 1 ) 2 b 1 = 0 . 5 Exercise 3 ( Regression without any regressor ) Suppose you are given the model Y i = 1 + u i . Use OLS to find the estimator of....
View Full
Document
 Fall '08
 Staff
 Econometrics

Click to edit the document details