Re: st: Binary choice model: Logit using IVs

[urbain thierry YOGO <yogout@gmail.com>]

you can run a twostep regression by hand using the predicted variable
in the second step. But as you said, you have to correct the standard
error. But remember the first step cannot be an OLS as far as you need
a non linear projection.

2012/4/3, Rüdiger Vollmeier <ruediger.vollmeier@googlemail.com>:
> Dear statalistusers,
>
> I read former posts stating that there was no (user-written-) command
> to estimate a simple binary choice model using Logistic regressions
> and instrumental variables (in case of endogenous regressors) (I guess
> nobody has come up with such a command since the problems were
> discussed (?)).
>
> My problem is as follows: I would like to estimate an Logit regression
> using IVs because all the regressions without IVs in my paper do rely
> on the logistic distribution of error terms.
>
> 1. Hence, I would like to know how you typically conduct a Logit using IVs?
> 2. Would it be possible to run a 2SLS "by hand" (OLS of endogenous
> regressors) and then Logit on the predicted values? (How) would I
> correct the standard errors of the parameter estimates in this case?
> *
> *   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/
>


-- 
*Urbain Thierry YOGO
P.h.D candidate in Economics*

*
*   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/