Statalist The Stata Listserver


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

st: RE: multi-level modeling with multiple imputed data in STATA


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: multi-level modeling with multiple imputed data in STATA
Date   Thu, 15 Mar 2007 18:26:21 +0100

--- tdavis7@emory.edu wrote:
> However, it doesn't appear that Stata supports multi-level modeling
> with any of the imputed data commands (mifit for example).  It also
> does not appear that gllamm or xtmixed support multiple imputed data. 
> Am I correct?

Yes, but this is not a big problem, since you can do it by hand. You 
can find the necessary formula's at: 
http://www.stat.psu.edu/~jls/mifaq.html#howto

Basically the point estimates are the mean estimates over the different
datasets, and the standard errors are the square root of the mean 
variances plus a simple to compute correction factor.

> I would think that Stata can combine the variables from the imputed
> data sets into an average data set?  Is this correct?  If so, is this
> even prudent?

This would be a bad idea. The whole point of multiple imputation is that
you get multiple datasets, and multiple estimates. It is the differences
between the estimates that tells us about the uncertainty due to 
imputation.

Hope this helps,
Maarten


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology 
Vrije Universiteit Amsterdam 
Boelelaan 1081 
1081 HV Amsterdam 
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434 

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


 

*
*   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