Hi Hans.
I had a somewhat similar problem a while ago and I got suggestions to
have a look at -biprobit-.
Here's my question, just scan the thread:
http://www.stata.com/statalist/archive/2004-12/msg00315.html
Best regards,
Guido
statalist-digest wrote:
----------------------------------------------------------------------
Date: Mon, 11 Jul 2005 11:23:14 +0200
From: "Hans J. Baumgartner" <[email protected]>
Subject: st: binary response model, where one regressor is binary and endogenous
Dear Statalist,
I would like to estimate a binary response model, where one regressor is
binary and endogenous. That is,
y1 = 1[d1*z1 + a1*y2 + u1 > 0]
y2 = 1[d2*z v2 > 0].
(see also Wooldridge 2002, p.477-8)
Is there a ready to use maximum likelihood estimator available to approach
this problem?
Thanks for any suggestions.
Hans
++++++++++++++++++++++++++++++++
Hans J. Baumgartner
Dept. Public Economics
DIW Berlin
German Institut for Economic Research
phone: +49/30/89789-307
fax : +49/30/89789-114
URL: www.diw.de
www.hansbaumgartner.de
++++++++++++++++++++++++++++++++
--
****************************************************************************
Dr. Guido Heineck
�sterreichisches Institut f�r Familienforschung
- Abteilung f�r Forschung & Entwicklung
Austrian Institute for Family Studies (�IF)
- Department of Research & Development
http://www.oif.ac.at/
e-mail: [email protected]
tel.: +43-1-535 14 54-23
fax: +43-1-535 14 55
address: A-1010 Wien / Vienna, Gonzagagasse 19/8
****************************************************************************
see also: http://www.8ung.at/heineck
*
* 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/
