[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE:st: SUR and a system of logit models

From   "Stephen P. Jenkins" <>
To   <>
Subject   RE:st: SUR and a system of logit models
Date   Sat, 26 Apr 2008 11:35:57 +0100

> ------------------------------
> Date: Fri, 25 Apr 2008 16:53:26 -0500
> From: "F. Gao" <>
> Subject: st: SUR and a system of logit models
> Dear all,
> I could not find reference on the following. Your help is 
> greatly appreciated.
> What to do if I have a system of logit models (binary dependent
> variables) and wish to do SUR regression?
> eg: Pr(A=1)=f(x,y), Pr(B=1)=f(x,z)
> ------------------------------

There is no such approach that I am aware of, as there is no
multivariate logistic distribution.

On the other hand, if you switch to probit link function, you can
exploit the multivariate normal distribution.  If you have only 2
dependent variables, as in your example, see -biprobit-.  If you have 3
or more dependent variables, see -mvprobit- (on SSC and SJ; use -findit-
to get latest version of the code). The article from SJ 3(3) 2003 about
the first version of the program is downloadable from SJ website.

Professor Stephen P. Jenkins <>
Director, Institute for Social and Economic Research
University of Essex, Colchester CO4 3SQ, U.K.
Tel: +44 1206 873374.  Fax: +44 1206 873151.  
Survival Analysis using Stata: 
Downloadable papers and software:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index