Dear Statalisters,
I am estimating a dynamic panel with -xtabond2- and suspect that one of the
regressors might be endogenous. To test for the endogeneity of the
regressor, I followed the procedure suggested by Mark Schaffer
(http://www.stata.com/statalist/archive/2004-07/msg00101.html), i.e. I first
estimated the regression treating the regressor as exogenous (which gives me
the Sargan statistic S0) and then I estimated the regression again treating
the regressor as endogenous (which gives me the other Sargan statistic S1).
Everything looks ok in those regressions. The problem is that the C test
would then be calculated as S0-S1. Here, I get a negative number (-0.01) and
since the statistic is distributed as chi-square, I have a problem. Does
anyone have a suggestion what to do about this?
Thanks in advance,
Cordula
*
* 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/
