Re: st: Heckman model and endogeneity

["Yu(Sophia) Zhang" <sophia.zhangyu@gmail.com>]

Hi Anik,

  I am doing the similar thing. But my endogenous variable is only in the
second stage, not in the selection equation. And I am using multinomial
logit model in the first stage. I think it works in my case. I am waiting
for confirmation from some faculty. Hope this information helps.

  Yu

On Fri, Nov 19, 2010 at 5:46 PM, ASIF REZA ANIK <anikbd1979@gmail.com> wrote:
>
> Hi All,
>
> I want to do a heckman model. The problem is that I assume one of my
> independent variable has endogeneity problem.
>
> The model is as follows:
>
> y=f(x1, x2, x3)
>
> if my selection criteria is s, the syntax for stata will be :
>
> heckman y x1 x2 x3, (select=s x1 x2 x3)
>
> But if I assume x3 has endogeneity problem, then what should I do. Is
> it okay if I do the followings:
>
> regress x3 x1 x2 x4 x5 ..................(x4 and x5 are instruments)
> predict y*
> and in the final model I replace x3 by y*, and the syntax is:
>
> heckman y x1 x2 y*, (select=s x1 x2 y*)
>
> Is anybody kind enough to tell me whether I am correct or not? I am
> new in the issues of Heckman model and endogeneity. So confused.
> Looking for your expert opinions.
>
> regards
>
> Anik
>
>
> --
> Asif Reza Anik
> Ph.D. Student
> Department of Project and Regional Planning
> Institute of Farm and Agribusiness Management
> Faculty of Agricultural Sciences, Nutritional Sciences and
> Environmental Management
> Justus-Liebig-Universität Giessen
> Senckenbergstr.3, D-35390,
> Giessen, Germany.
> Tel: +49 (0) 176 811 25989 (Mob.)
> E-Mail: Asif.R.Anik@agrar.uni-giessen.de
> anikbd1979@gmail.com
> http://www.uni-giessen.de/cms/fbz/fb09/institute/ibae/Regionalplanung/daad/Research/Anik
>
> *
> *   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/



--
Yu Zhang
PhD Candidate
Economics and Education
Teachers College, Columbia University

*
*   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/