Statalist


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

RE: st: xtlogit is slow


From   jverkuilen <jverkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: xtlogit is slow
Date   Sun, 6 Jul 2008 16:00:30 -0400

SAS GENMOD is using GEE. Intergration is only needed for random effects (re) models so altering the number of integration points shuld change nothing.

You can get a GEE in -xtlogit- by using the population averaged option, which should be pretty quick. You could also try -xtgee-. I am really surprised Stata is taking so long and suspect you're asking for the re solution. If not, something's funny. 

-----Original Message-----
From: "Mitchell F. Berman" <mfb1@columbia.edu>
To: "Stata List Server" <statalist@hsphsun2.harvard.edu>
Sent: 7/6/2008 7:15 AM
Subject: st: xtlogit is slow

Stata List Users:

I've noticed that xtlogit is slow compared to SAS genmod to get logistic 
regressions using an exchangeable correlation model.  For large data 
sets (300,000 rows-15 variables) xtlogit takes 45 minutes, SAS takes 1.5 
minutes.  The results are identical.

Is there any way to speed this up?  There was a post a few years back 
that suggested changing the number of points used in the approximation 
to the integral.

Does anyone know how to do that or other ways to speed up xtlogit?

Thanks in advance.

Mitchell Berman
Columbia University
New York
*
*   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