RE: st: xtlogit is slow

["Mitchell F. Berman" <mfb1@columbia.edu>]

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/