Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: re: 2-SLS with no covariates in 1st stage


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: re: 2-SLS with no covariates in 1st stage
Date   Mon, 3 Sep 2007 07:52:04 -0400

Eva said

ivreg in STATA fits a linear regression model using instrumental variables (or
two-stage least squares) of depvar on varlist1 and varlist2, using varlist_iv
(along with varlist1) as instruments for varlist2.Is there any option in ivreg
command for not using varlist1, along with varlist_iv, as instruments (I want
to keep out from the first-stage of the 2-SLS the varlist1)?


No. This is a FAQ (see the Statalist Archives, Mark Schaffer's postings in particular). The instrumental variables estimator requires that you include all exogenous variables in the Z matrix. The elements of varlist1 are serving as their own instruments in the 'first stage' regression, and excluding them would be equivalent to saying that they are orthogonal to the elements of varlist2. As you can readily verify that the correlations between varlist1 and varlist2 are nonzero in all but very special cases, the elements of varlist1 must be included in the 'first stage' regression to ensure consistency.


Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index