Kai,
Date sent: Wed, 13 Nov 2002 16:35:21 +0100
From: Kai Kirchesch <[email protected]>
To: [email protected]
Subject: st: Sargan's difference test
Send reply to: [email protected]
> I am using the xtabond command for GMM estimation. However, the results
> only report the Sargan statistic.
> Is there any possibility to perform the Sargan's difference test with Stata?
> Thanks,
> Kai
If you are using xtabond, can you not estimate the constrained and
unconstrained versions of the equation and take the difference of the
two Sargan statistics?
This should be correct asymptotically, but you may run into the
problem of a negative test statistic, akin to the negative Hausman
test statistic that was discussed on the list last week. Even this
can probably be easily solved with the xtabond one-step Sargan
statistic by using the same error variance (presumably from the more
efficient estimation) for the two Sargan stats before taking the
difference.
--Mark
Prof. Mark E. Schaffer
Director
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
http://www.som.hw.ac.uk/cert
*
* 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/
