Re: st: RE: -anova- command fails with mildly complexx mode

[Philip Ender <ender97@gmail.com>]

David Airey wrote:

It's interesting that the equivalent -xtmixed- command doesn't crash,
so that's an alternative approach while Stata Corp tracks down and
fixed this -anova- related crash. Note the use of -anovalator- (by
Phil Ender) combined with the default indicator factor coding instead
of using effect coding and avoiding factor variable use. Your design
has two between subject variables treat and group and one within
subject variable time.

xtmixed y i.treat##i.group##i.time || isub:
anovalator treat group time, main
anovalator treat group, two
anovalator treat time, two
anovalator group time, two
anovalator treat group time, 3way

--------------------------------

The only thing I would suggest is to use the -fratio- option with the
-anovalator- commands to get the results
scaled as F-ratios instead of chi-square.

Version 1.9 of -anovalator- is now available from the ATS Stat group. Just type:

. findit anovalator

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/