I wanted to follow up on this request: it appears that -gllamm- is a
random effects model, not a generalized estimating equation model. If
xtgee supported a -cumlogit- or -ologit- link function, I'd be home
free. Is there an extension available for xtgee or something similar?
Thank you!
Laura
>>> [email protected] 09/03/03 02:08PM >>>
On Wed, 3 Sep 2003, Laura Damschroder wrote:
> Hello -
>
> I am finally happily migrating to Stata from using a combination of
SAS
> and SPSS and have succeeded in replicating most of the models and
> analyses I have done in the past in Stata. However, there is one
model
> I cannot find any documentation for in Stata: what if I have an
ordinal
> dependent variable with repeated measures (a within-subject measure)
and
> a categorical independent variable along with the variable
indicating
> the "panel" (scenario in this case). In SAS, I coded this using:
>
> proc genmod data=geeanal.age DESC;
> class id educat scenario;
> model preference = educat scenario/
> dist=mult link=cumlogit type3;
> repeated subject=id / corrw type=indep;
>
> Does anyone know how to replicate this model in Stata?
Try -findit gllamm- and -findit reoprob-
Stephen
*
* 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/
