This preview shows pages 1–5. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Economics 326 Methods of Empirical Research in Economics Lecture 19: Linear regression without strong exogeneity Vadim Marmer University of British Columbia May 5, 2010 Strong exogeneity and the conditional expectation function (CEF) I Consider the linear regression model Y i = + 1 X i + U i . I When the errors are strongly exogenous , i.e. E ( U i j X i ) = 0, the linear regression model de&nes the CEF of Y conditional on X : CEF Y ( X i ) & E ( Y i j X i ) = E ( + 1 X i + U i j X i ) = + 1 X i + E ( U i j X i ) = + 1 X i . 1/9 Weak exogeneity Y i = + 1 X i + U i , EU i = I Suppose the errors are only weakly exogenous: E ( U i X i ) = . I In this case, CEF Y ( X i ) 6 = + 1 X i . I Question: What does the econometrician estimates when he runs a linear regression and the regressors are not strongly exogenous? 2/9 Linear regression as a misspeci&ed CEF I Suppose that E ( Y i j X i ) = g ( X i ) , where g is some unknown nonlinear function. Thus, the truefunction....
View Full
Document
 Spring '10
 whisler
 Economics

Click to edit the document details