Dear Statalist Users,
I've been looking around in the archives but haven't found what I'm
looking for. I would very much appreciate your help.
I have a binary dependent variable so I'm trying estimate a probit model.
I have ten independent variables, *four* of which are possibly endogenous
so I would like to instrument these. I have panel data and I would like to
cluster my standard errors if possible.
Now there is an existing procedure, ivprob, but I can only have a single
endogenous variable it seems. So I have a couple of questions:
(1) Has anyone tried to estimate a probit model with many endogenous
variables and how did you implement this in Stata?
(2) Does anyone know of a reference that estimates probit with many
endogenous variables?
(3) If I were to write my own code, it should be in principle be
straightforward to extend the single-endogenous-variable case to the
multiple-variables case, but the covariance matrix of the errors (in my
case, 5 x 5) becomes very cumbersome to work with. Any hints from
experienced users about possible ways of simplification?
I really appreciate your help with this. Thank you very much.
Jason.
*
* 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/
