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

st: RE: identifying perfect outcome predictor

From   "Newson, Roger B" <>
To   <>
Subject   st: RE: identifying perfect outcome predictor
Date   Mon, 5 May 2008 19:13:46 +0100

I personally use -glm- (with the options -link(logit) family(bin)-)
instead of -logit-. That way, the offending parameters are allowed to
"converge" to plus or minus infinity without an error message. And the
guilty parameters are then displayed for all to read.

I hope this helps.


Roger B Newson
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Web page:
Departmental Web page:

Opinions expressed are those of the author, not of the institution.

-----Original Message-----
[] On Behalf Of Feiveson,
Alan H. (JSC-SK311)
Sent: 05 May 2008 18:34
Subject: st: identifying perfect outcome predictor

Hi - I am running logistic regression on simulated data sets. However
sometimes one of the explanatory variables completely separates the
outcome variable and I get a message such as 

outcome = X2 <= .1955861 predicts data perfectly

Presumably if I get a return code of 2000, I know this has occurred -
but is there information in logit postestimation to tell which variable
gives perfect separation (as opposed to checking each one "by hand")?


Al Feiveson

*   For searches and help try:

*   For searches and help try:

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