Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: multilevel ordered logistic, adjusted prob


From   "INAGAMI,SANAE " <sinagami@ucla.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: multilevel ordered logistic, adjusted prob
Date   Wed, 9 Mar 2005 10:21:34 -0800

I am a beginner.  I've looked at a lot of stuff on the web and can't make
heads or tails.

I ran gllamm on an ordered logistic model which had two levels: census
tracts and individual.

I would now like to calculate conditional probalities of an ordered
outcome based on a categorical covariate being 1 or 0.

i.e.  Outcome of self-rated health if everyone were married or everyone
were not married.  Not outcome of an individual but an average outcome if
everyone were married.  Then I'd like to create a ratio for a "relative
risk" of sorts.  An outcome of very poor health if everyone were married
vs an outcome of very poor health if everyone were not married.
Then I'd like to develop a confidence interval around this ratio.

I can do this in stata if it were on one level.  I used gllamm because of
the multilevel nature.  And I think I can use gllapred for what I want,
but having read the gllamm manual and the statalist conversations, I
still can't figure out how to set up the syntax.

Thanks.
Sanae
*
*   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/



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