Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

re: st: Kenward-Roger method in Stata?


From   "Airey, David C" <david.airey@Vanderbilt.Edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   re: st: Kenward-Roger method in Stata?
Date   Fri, 4 Jun 2010 08:40:30 -0500

.

This is a useful summary article of the denominator df in mixed models:

http://www2.sas.com/proceedings/sugi26/p262-26.pdf

The article concludes, "The approximate ddf methods come closer to serving as automatic methods than the exact methods. However, complexity of the covariance structure, sample size, and imbalance affect the performance of both approximate ddf methods."

Whereas the KR method is an option in SAS PROC MIXED, it is the default setting in SAS JMP.

The method is not cited in the -xtmixed- references, although some of those citations are not primary sources, and the issue is not discussed in the -xtmixed- in the manual, or I missed it just now.

Kenward & Roger (1997), BIOMETRICS 53, 983-997 "Small Sample Inference for Fixed Effects from Restricted Maximum Likelihood" 

-Dave
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index