Statalist


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

st: How to check exogeneity when I use "ivprobit" together with robust and cluster


From   "Sun, Yan (IFPRI)" <Y.SUN@cgiar.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: How to check exogeneity when I use "ivprobit" together with robust and cluster
Date   Mon, 24 Sep 2007 16:45:55 -0400

Hi, there,

My dependent variable is a dummy, and I have endogenous problem, so I
use "ivprobit" command together with robust and cluster. After checking
over identification test and relevance test (I have to thank professor
Baum for all his great help), I now try to test the exogeneity of the
potential endogenous variable . I try to use "probexog" for doing
Smith-Blundell test, but this command does not allow robust option. I
also try to use hauseman test and use Wooldridge method manually, but I
am not sure these two tests are suitable for a dummy dependent variable.
I notice "ivprobit" (with robust and cluster) reports wald test of
exogeneity (chi2 square) after the regression, does this null hypothese
is the potential endogenous variable is exogenous? 

Another question is when I do "ivprobit" (with robust and cluster), the
wald chi2 in the beginning of the main regression is ".", so the prob >
chi2 is also ".", but the main regression results are fine. What is the
problem? Thank you very much. 

Yan Sun
IFPRI



*
*   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