Dear Statalisters,
I have the following model to be estimated for a panel: y = a +bx + cz
+ d(x*z) + (....),
where x is endogenous. Therefore, I estimated the model using
XTIVREG2. However, I am concerned that because x*z is the interaction
between an endogenous variable and an exogenous variable, it is itself
an endogenous variable. Obviously, to find a good instrument for this
interaction variable is probably more complex than usual. Billet, Kim
and Mauer (2007) suggests that estimating the model via nonlinear GMM
would address the endogeneity of the interaction variable.
I would appreciate if somebody can suggest how I could do nonlinear
GMM within XTIVREG2 or STATA in general.
Thanks,
Erasmo
Reference
MATTHEW T. BILLETT, TAO-HSIEN DOLLY KING, and DAVID C. MAUER, Growth
Opportunities and the Choice of Leverage, Debt Maturity, and
Covenants, THE JOURNAL OF FINANCE • VOL. LXII, NO. 2 • APRIL 2007
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
