Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Austin Nichols <austinnichols@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: logit fixed effects models with weights |
Date | Wed, 5 May 2010 11:03:23 -0400 |
Christina Gibson-Davis <cgibson@duke.edu> : Do weights vary at the individual level because there are multiple waves of survey data, and waves after the first are adjusted for attrition? In that case, I would use the first-wave weight for each individual: bys id (wave): g w1=weight[1] clogit y x [pw=w1], gr(id) vce(cl id) On Wed, May 5, 2010 at 10:48 AM, Christina Gibson-Davis <cgibson@duke.edu> wrote: > Hi, > I am trying to estimate a logit fixed effects models with weights. -xtlogit- does not support either fweights or pweights. I am working with a dataset with more than 18,000 individuals, so it is not possible to include a dummy var for each individual. -clogit- supports weights, but they must be constant within group, and my weights vary at the individual level. My outcome is binary so -areg- will not work. > > Are there any suggestions as to how to estimate this model? > > Many thanks in advance. > > Christina * * 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/