Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Estimating Simultaneous Equations Models with GMM 3SLS estimator


From   Helene Ehrhart <Helene.Ehrhart@u-clermont1.fr>
To   statalist@hsphsun2.harvard.edu
Subject   st: Estimating Simultaneous Equations Models with GMM 3SLS estimator
Date   Thu, 12 Mar 2009 11:34:17 +0100

Hello,

I am estimating a two-equation linear simultaneous equations model where the 'exogenous' variable in one equation might be correlated with the error in the other equation. To estimate the model I used traditional 3SLS with the reg3 command in Stata. However, I read in Woolridge' Manual (Econometric Analysis of Cross Section and Panel Data, chapter 9.6) that "the traditional 3SLS estimator is generally valid only when instruments are uncorrelated with all errors". It is the GMM 3SLS estimator that must rather be used to produce consistent estimations.

In the reg3 command there is no GMM option, so I am wondering which command I should use to estimate my simultaneous equation model with the GMM 3SLS estimator.

Thank you.

Helene Ehrhart
CERDI, France.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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