This sounds like a nice option. Does this imply that I can run a
stepwise regression or LARS by using cmdok?
Tony
Peter A. Lachenbruch
Department of Public Health
Oregon State University
Corvallis, OR 97330
Phone: 541-737-3832
FAX: 541-737-4001
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Yulia
Marchenko, StataCorp LP
Sent: Sunday, August 09, 2009 12:14 PM
To: [email protected]
Subject: Re: st: Factor variables and mim
Fred Wolfe <[email protected]> pointed out that although -mi
estimate- supports factor variables, it does not support -xtmixed-:
> Yes, -mim- does not accept factor variables - too bad. and -mi
estimate-
> doesn't work with xtmixed or mfp
> mim works with xtmixed.
>
> From my perspective, if you want to use multiply imputed data sets and
> -xtmixed-, it appears that -ice- and -mim- are what must be used, but
> without factor variables - something that could be done in Stata 10.1.
Fred is right that -mi estimate- does not yet officially support
-xtmixed-.
However, -mi estimate- provides a way of obtaining results from
virtually any
estimation command via option -cmdok-. For example, you can type:
. mi estimate, cmdok: xtmixed y x1 i.x2 || group:
to obtain MI results from -xtmixed- with factor variables.
--Yulia
[email protected]
*
* 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/
*
* 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/
