Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: predict in xtlogit re


From   "Scott Merryman" <merryman@icehouse.net>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: predict in xtlogit re
Date   Fri, 19 Jul 2002 15:27:21 -0600

The default for predict  in xtlogit ,re is the fitted values not the
predicted probabilities; for that you want to use predict varname, pu0

The default in xtlogit, fe is predicted probability conditional on one
positive outcome within group.

See  [R] xtlogit.

Scott Merryman

----- Original Message -----
From: "Botelho, Delane" <delane.botelho@marshall.usc.edu>
To: <statalist@hsphsun2.harvard.edu>
Sent: Friday, July 19, 2002 3:16 PM
Subject: st: predict in xtlogit re


> Hi
>
> When I run xtlogit (random effect) and run predict, I get predictions
> that are greater than 1. The default for RE is xb.
>
> When I run xtlogit FE (fixed effect) and I run predict, I get predicted
> probabilities that are between 0 and 1. The default for FE is pc1.
>
> What is the logic? Why not having always predicted probabilities
> (between 0 and 1) if logit is a probability model?
>
> Thanks
>
> Delane Botelho
>
>
> *
> *   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index