Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: longitudinal ordinal regression


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: longitudinal ordinal regression
Date   Wed, 3 Jun 2009 15:27:29 +0100

Can you identify any literature that defines covariance for ordinal (graded) responses? I don't usually align myself with "level of measurement" arguments, but I would be surprised if the models you think you want are even defined in the literature. I'd be happy for anyone to expose my enormous ignorance on this detail. 

(I am confident that Stata is correctly so spelled, but that of course is a detail.) 

Nick 
n.j.cox@durham.ac.uk 

Shubhabrata Mukherjee

    I-emailed statalist about a month back regarding longitudinal ordinal regression in STATA (as I couldn't find anything such as `xtologit').

     A few people recommended that I should use GLLAMM or ologit (with `cluster id' option). But there is no option to specify the type of covariance structure (unstructured, ar1 etc.) on either of them if I am not wrong. I think the cluster(id) option for ologit (or i(id) for GLLAMM) are just nesting the observations by their respective ids.

    I was wondering if anyone can point out some other models which could take care of this problem (specifying covariance structure in a longitudinal OLR)?

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index