Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: RE: Re: RE: re: RM ANOVA, was SPSS vs. Stata
From
Philip Ender <[email protected]>
To
[email protected]
Subject
Re: st: RE: Re: RE: re: RM ANOVA, was SPSS vs. Stata
Date
Tue, 3 Aug 2010 09:46:17 -0700
"Feiveson, Alan H. (JSC-SK311)" <[email protected]> wrote:
Phil, and others
For larger data sets with high imbalance I don't think there's much
doubt that using a mixed model is more flexible and less biased than
rpm anova with complete observations only. But for small sample sizes,
using infinite degrees of freedom for the denominators (i.e.
chi-square statistics rather than F) also creates bias in the
inference. What would be nice is to have some way to calculate
approximate denominator degrees of freedom after obtaining the
pseud0-F statistics with -xtmixed- and -test-.
Al Feiveson
----------
Al-
I've had some discussions on this topic with Vince Wiggins. He claims
that regardlesss of what SAS is doing there isn't a statistically
defensible of determining the denominator degrees of freedom. For
linear models rescaling chi2 to F yields a correct F-ratio but without
denominator df its difficult to get a p-value. I have looked back at
my old BMD manual for the 05V procedure and note that it only produces
z-tests and chi2 although it was specifically designed for unbalanced
repeated measures. So, while not prefect, I do think the F-ratio I
obtained in my -xtmixed- example is less biased than than the one from
the complete case analysis in SPSS.
Phil
--
Phil Ender
UCLA Statistical Consulting Group
*
* 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/