# st: RE: RE: RE: 2SLS, logistic regression and more (problems...)

 From Tatjana Oberd�rster <[email protected]> To "[email protected]" <[email protected]> Subject st: RE: RE: RE: 2SLS, logistic regression and more (problems...) Date Thu, 13 Dec 2007 13:49:46 +0100

```Thanks a lot for your comment!

I�ve just come across another question, not directly related to Stata, but I have not yet found any hints in the textbooks.

Is it possible to use ranked data in 2SLS estimations? My data is ordinal scaled and comprises nonlinear relationships. In accordance to existing literature in my field (accounting) I calculated ranks and so far used them in ordinary OLS-estimations. I am just wondering if I could use 2SLS in this case since I need to incorporate instrument variables due to an endogenous explanatory variable.

Tatjana

-----Urspr�ngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von Corey Phelps
Gesendet: Samstag, 8. Dezember 2007 22:26
An: [email protected]
Betreff: st: RE: RE: 2SLS, logistic regression and more (problems...)

It's unclear to me why you want to estimate the first stage using logistic
regression. I am assuming your enogenous variable is dichotomous. If so, an
you are only interested in obtaining a consistent estimate in the second
stage, you can use linear regression for the first stage. Estimating the
first stage using probit or logit is unnecessary because in 2SLS the
consistency of the estimates in the second stage are not dependent upon
specifying the correct functional form in the first stage (Kelejian, 1971).

Kelejian, H. 1971. Two-stage least squares and econometric systems linear in
parameters but nonlinear in endogenous variables. Journal of the American
Statistical Association, 66: 373-374.

Corey Phelps, PhD
Asst. Professor, Management & Organization
Neal & Jan Dempsey Faculty Fellow
Michael G. Foster School of Business
University of Washington
Box 353200
Seattle, WA 98195
(206) 543-6579
www.coreyphelps.com

> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Verkuilen, Jay
> Sent: Saturday, December 08, 2007 1:02 PM
> To: [email protected]
> Subject: st: RE: 2SLS, logistic regression and more (problems...)
>
> Tatjana Oberd�rster wrote:
>
> >>I'd like to do a 2SLS estimation where the first-stage
> regression is Logistic. [snip] <<
>
> that can be found at http://www.gllamm.org.
>
> Jay
> --
> J. Verkuilen
> Assistant Professor of Educational Psychology
> City University of New York-Graduate Center
> 365 Fifth Ave.
> New York, NY 10016
> Email: [email protected]
> Office: (212) 817-8286
> FAX: (212) 817-1516
> Cell: (217) 390-4609
>
>
> *
> *   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/
>

*
*   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/

*
*   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/
```