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

st: matched case logit with random effects


From   "Jacqui Frair" <jfrair@ualberta.ca>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: matched case logit with random effects
Date   Sun, 3 Apr 2005 11:50:51 -0600

Hello Statalist,

I have a question regarding the integration of random effects into a
conditional (fixed) effects logistic model.  I am comparing attributes of
locations used by elk to locations they were not present in.  So used sites
are coded as 1 and available sites (randomly selected area near each used
site) are coded as 0.  This is actually a 1:n matched case design in that I
have 20 randomly selection locations to each used location. I compare use to
availability by a series of continuous as well as categorical environmental
covariates.  Since I am studying multiple individuals I would like to
include a random effect of individual and in particular I would like to
assess if individuals differ in their magnitude of association with
different variables (due to environmental gradients).  Thus, I would like to
include random coefficients.  I am experienced with XTLOGIT and GLLAMM to
some degree but haven't quite figured out how to accomplish a conditional
logistic model with random effects.  It has been recommended to me that I
might accomplish a random "intercept" model using a stratified Cox model
where the strata are each set of used and available points at each time
step.  I have used STCOX for survival analyses and this alternative use
isn't clear to me, if anyone has any experience with this can you comment?
Even if it does work though I don't see how random coefficients could be
included, at least in the Stata version.  Does anyone see a way to use
GLLAMM to accomplish a 1:n matched case logistic model?  I thought perhaps
through the mlogit link but haven't gotten my head around it yet.  I'm
hoping someone on this list may have experience and can give me some
insight!

Thank you.

Sincerely,
Jacqui

********************************************
Jacqueline L. Frair
Ph.D. Candidate
Department of Biological Sciences
University of Alberta
Edmonton, Alberta T6G 2E9
PHONE:  780-492-0083
FAX:  780-492-9234
EMAIL:  jfrair@ualberta.ca
********************************************

*
*   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