Statalist The Stata Listserver


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

st: separation problem


From   "Matthew Henry Kroenig" <kroenig@uclink.berkeley.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: separation problem
Date   Wed, 29 Mar 2006 16:26:30 -0800

Hi,

I am doing logit analysis and I have a separation problem--one of my model's covariates perfectly predicts the binary outcome. Stata ommits the variable and drops observations when separation is present. According to Christopher Zorn [ "A Solution to Separation in Binary Response Models" Political Analysis, 13:2 (2005)], I can solve the separation problem by estimating a penalized-likelihood model using, for example, Firth's bias-reduced logistic regression (brlr). As far as I can tell, Firth's own brlr package is for R only. Is there a version of brlr for Stata? How can I estimate a penalized-likelihood model in Stata?

Thanks in advance for your time.

Best,

Matt Kroenig
National Science Foundation IGERT Fellow
Department of Political Science
210 Barrows Hall
UC Berkeley
Berkeley, CA 94720
USA

(510) 499-1575
kroenig@berkeley.edu
*
* 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/




© Copyright 1996–2020 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index