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

st: re: 2sls with multiple endogenous variables

From   Kit Baum <[email protected]>
To   [email protected]
Subject   st: re: 2sls with multiple endogenous variables
Date   Fri, 14 Dec 2007 18:16:54 -0500

Ana said

I am trying to estimate a 2SLS model with two
endogenous variables (y2, y3) which have different
exogenous explanatory variables. That is,
y2=f(z1,z2) and y3=f(z3,z4,z5). Is there any Stata
command that will allow the estimation of this model?

No, this is a FAQ. 2SLS requires that you regress each endog on all of the instruments, included and exlcuded.

If you specify a full mulitple-equation model (e.g. 3SLS) you can specify the exclusion restrictions that apply to the RHS endogenous variables (as they are then LHS in each of their own equations) but in a 2SLS/IV context, it cannot be done. Consistent estimation of 2SLS requires that all instruments appear in each 'first stage' regression.

See the Stata FAQs.

Kit Baum, Boston College Economics and DIW Berlin
An Introduction to Modern Econometrics Using Stata:

* For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index