Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# RE: st: re: How to correct standard errors of a 2sls performed by

 From Anne-Sophie Bergerès To Subject RE: st: re: How to correct standard errors of a 2sls performed by Date Wed, 17 Feb 2010 23:07:03 +0100

```Thank you for your answers.

However can i use cmp if i want to do a logistic in the first stage ?

I have this type of model

y1 = bo+ b1X1 + b2X2 + b3X3+ ...+ e1
y2 = b0 + b4y1 +....+ e2

where y1 is binary. So i have a dummy endogeneous variable and my model is over-identified.

I need to run my 2sls with a logistic in the first step and not an ols (as in ivreg2)

So i have decided to do it by hand.

1- run logistic of y1 on all exogenous variables of the models + instruments
2- save residuals
3- run y2 on variables in the equation + res.

But i need to correct the standard errors in order to do an hausman et sargan test also by hand.

Cmp would do that ? it does not seem to take into account logistic.

Or can i use vce(bootstrap) ?

I am really not sure of what to do and to get the logic behind corrected standard error.

I thank you very much in advance (again).

Regards

Anne-Sophie

> From: baum@bc.edu
> Subject: st: re: How to correct standard errors of a 2sls performed by
> Date: Fri, 5 Feb 2010 19:56:56 -0500
> To: statalist@hsphsun2.harvard.edu
>
> <>
> Annie-Sophie said
>
> In fact i use a logistic in the first step and ols in the second. I have one endogeneous binary variable. Therefore should i use bootstrap ? Is it sufficient to use vce (bootstrap ) ?
>
> -findit cmp-
>
> Kit Baum | Boston College Economics & DIW Berlin | http://ideas.repec.org/e/pba1.html
> An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html
> An Introduction to Modern Econometrics Using Stata | http://www.stata-press.com/books/imeus.html
>
>
> *
> * 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/
_________________________________________________________________
Tchattez en direct en en vidéo avec vos amis !
http://www.windowslive.fr/messenger/
*
*   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/
```