Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: IVTOBIT with 2 PROBITS on first stage...


From   "Diana Fletschner" <fletschn@u.washington.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: IVTOBIT with 2 PROBITS on first stage...
Date   Wed, 22 Aug 2007 11:08:25 -0700

Hello--

I am trying to estimate a TOBIT model with two endogenous binary variables. That is, a model like:
y = b0 + b1 X1 + b2 X2 + ... + u
where y has an upper bound of 1 and X1 and X2 are either 0 or 1. I would like to instrument X1 and X2 using either two PROBIT models (or even better, a BIVARIATE PROBIT model since the error terms for X1 and X2 are likely to be correlated).

I have tried using IVTOBIT but cannot get results:
* if I use the FIML option, it tells me the function is not concave.
* if I use the two step option (which I'd prefer), I encounter two problems:
1) the first stage uses OLS not PROBITs; and,
2) I cannot use pweights, something I need given my sample.

If you have any suggestions on how to approach this problem, I would love to hear from you. Any help would be much appreciated.

Thanks!
Diana


_________________________________
Diana Fletschner
Assistant Professor
Daniel J. Evans School of Public Affairs
University of Washington
(206) 616-1297

*
* 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/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index