st: two endogenous varaibles with xtabond2

["Lovisa Persson" <lovisa.persson@nek.uu.se>]

Hello users,

I am estimating a model using xtabond2 
It looks the following:

xtabond2 costs revenue controls, gmm(revenue, laglimits(2 3)) iv(controls)
robust noleveleq artests(3)

I am estimating the effect of change in revenues on the change in costs...
I am treating my controls as exogenous.

Now I would like to see if the estimate on "revenues" depend on whether you
belong to group "south", so I have a dummy for this group.
Now you can do this by splitting the sample and do the estimations
separately, but I would like to try to include an interaction variable
instead.

Interaction=south*revenue

And this interaction variable is then also endogenous since "revenue" is.
How do I proceed to estimate this model using xtabond2, with two endogenous
variables?

Thank you!


Lovisa Persson
Ph.D. Student 
Department of Economics 
Uppsala University 
E-mail: Lovisa.Persson@nek.uu.se


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/