Re: st: xtlogit is slow

[David Airey <david.airey@vanderbilt.edu>]

.

Does dropping the data to the estimation sample and removing the "if"  
change the timing? There has been a lot of discussion about "if"  
slowing Stata down. I doubt this is your problem.

-Dave

On Jul 7, 2008, at 8:54 AM, Mitchell F. Berman wrote:


> Thank you for the information that changing the integration points  
> won't help; I won't bother investigating that.
>
> The call to xtlogit that I'm using is:
>
> xi: xtlogit depvar i.catvar1 other_vars if rand <7, pa c(exc)
>
> the c(exc) option is unnecessary--- xtlogit ..., pa is equivalent to
> xtgee ..., family(binomial) link(logit) corr(exchangeable)
>
> I can't see what else I'm doing wrong, especially since the  
> identical data set runs quickly in SAS with results identical to  
> stata.  I'm guessing that the numeric method SAS uses for genmod is  
> just faster than whatever Stata is using for xtgee.
>
> Mitchell Berman
> Columbia University
>
>
>
>
> *
> *   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/