Hi everyone,
I too have problems using xtabond2. and I would very much appreciate your help
the following qusetion.
I am estimating a dynamic panel data model using One Step robust GMM System
estimator. I have a large data set with more than 50,000 observations and more
than 6,000 firms. the panel has three dimensiona: year, industry & country.
The problem I am facing is that the validity of instruments is rejected by the
Hansen J test of over-identifying restrictions (P-values =0). This is the case
when I include all variables that need to be instrumented in the gmm option of
xtabond2 command. But when I choose to include only one or two of them in gmm
option, I get P-values >0.
My question is as follows: is it correct not to include all endogenous/
pre-determined variables in the gmm style optionof xtabond2 as long as the
over-identifying restrictions are satisfied (ie as long as the number of
instruments used from outside of the current equation is greater than the number
of endogenous variables that need to be instrumented)
Thank you & best regards,
Dahlia Anwar El- Hawary
Consultant
Financial Sector Operations and Policy Department
World Bank
Tel: 202 473 5238
*
* 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/
