st: xtivreg with endogenous nonlinear regressor

["Jerg Gutmann" <jerg.gutmann@uni-hamburg.de>]

Hello everyone,

I would like to estimate a random-effects IV regression with panel data
using xtivreg. There are two types of models I am potentially interested in:

y = a + b*x1 + c*x1^2 + d*x2 + . + e

and

y = a + b*x1 + c*x1*x2 + d*x2 + . + e

In both cases x1 is endogenous and I have a suitable instrument z1 for it.
As I also have to instrument for either x1^2 or x1*x2, I also need a second
instrument.
In my understanding that could be an interaction of z1 with an exogenous
regressor from the structural equation.
Now my question is: Can I also use z1^2 as the second instrument in one or
both of these cases?

Best,
Jerg

*
*   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/