Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: xtmixed - fixed vs random effects


From   "Hillel Alpert" <HALPERT@hsph.harvard.edu>
To   "statalist" <statalist@hsphsun2.harvard.edu>
Subject   st: xtmixed - fixed vs random effects
Date   Thu, 28 Sep 2006 16:39:44 -0400

Could someone please advise with regard to specifying random and fixed variables in xtmixed?  My prior understanding is
that among other criteria, variables assumed to be measured without error (or that can have only a number of specific
values) would be fixed, and that random variables are assumed to be drawn from a larger population of values.

A continuous variable such as "weight" would seem to be random, but would it not also be designated as a fixed variable
in a mixed model in order to test the significance of its average effect on the outcome variable? A related question is,
if this is the case, when would the fixed effect be used to determine the significance and/or size of the effect of the
variable, and when would the random effect results or likelihood ratio test be used?

Hillel


*
*   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