Does
"two groups of clustered observations, but lots of observations per cluster"
mean id takes on two distinct values? And you have many obs in each
category? Just use logit with a dummy variable for id in that case.
On Sat, Nov 14, 2009 at 2:16 PM, Airey, David C
<[email protected]> wrote:
> .
>
> In logit models can one get faster results using weights if there are few covariate patterns? I thought I remembered reading that somewhere. If so, is it possible to do the same with xtlogit? So if you have the simplest of model, where you have two groups of clustered observations, but lots of observations per cluster, making estimation slow, can weights be used to speed up the estimation?
>
> xtlogit yn trt, i(id)
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
