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

Re: st: xtprobit problem


From   jlinhart@stata.com (Jean Marie Linhart, StataCorp LP)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: xtprobit problem
Date   Wed, 22 Jun 2005 15:13:13 -0500

"Salkov, Rick" <Rick.Salkov@eia.doe.gov> writes:

> I was trying to run a panel probit model but the program hung after
> 2 itterations of the full model.
>
> the data is large with 450,00 observations anf 240 variables, but
> using clogit, the model converges after 4 itterations.
>
> Is there anything I can do to get model results?

That is a *lot* of data. 

I managed to get my computer to practically grind to a halt with
450,000 observations and only 25 variables, so I think -xtprobit- is
not hung here, merely taking a really long time.  Even this reduced
model was taxing my patience, so I quit and timed four smaller models.
Based on the four smaller models and the assumption that computing
time increases linearly with the amount of data, I calculate it would
take my computer about a full 24 hour day to do the -xtprobit- on
450,000 observations and 240 variables.  (The model I started --
450,000 observations and 25 variables -- should take about 2-3 hours
to complete!)

I'd suggest running this overnight, if not over several nights.  If
waiting doesn't yield some progress, please write in to technical
support.  We will probably need details about your computer and data
in order to be of assistance.

--Jean Marie
jlinhart@stata.com
*
*   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