Statalist


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

st: distinguishing categorical from ordinal variables in xtmixed


From   "Olga Lyashevska" <olga.stata@herenstraat.nl>
To   statalist@hsphsun2.harvard.edu
Subject   st: distinguishing categorical from ordinal variables in xtmixed
Date   Thu, 24 Apr 2008 20:11:36 +0200 (CEST)

Dear All,

We are trying to use xtmixed as a mixed effect anova taking advantage of
maximum likelihood estimation.

In xtmixed Stata seems to interprete categorical (cardinal) fixed factors
(such as colour) as ordinals. How do we get Stata to treat these as a
simple categorical?

e.g.

xtmixed y colour || individual

where y is dependent continuous variable, colour is {red,green,blue}(coded
with an integer) and individual is a random factor (coded with an integer)

produces output in terms of regression coefficients etc. implying that
STATA interpreted the categoricals as ordinal variables. As though the
order was important, but of course it is not for categorical.


Thank you in advance,
Olga

*
*   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