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: Kenward-Roger method in Stata?
From
"Airey, David C" <[email protected]>
To
"[email protected]" <[email protected]>
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/