# Re: st: Explained variation / variance partitioning using xtmixed

 From Jeph Herrin To statalist@hsphsun2.harvard.edu Subject Re: st: Explained variation / variance partitioning using xtmixed Date Fri, 16 May 2008 10:18:10 -0400

Wouldn't you estimate the same model with and without
the covariates, and use the two estimates of sigma^2?

Patrick P. Phillips wrote:
Dear StataList,

Referring to the paper by Xu in Statistics in Medicine 2003,
"Measuring explained variation in linear mixed models"
(22:3527-3541), has this measure of explained variation been
implemented in Stata as a postestimation command to xtmixed or
anywhere else?

From xtmixed, Stata gives the residual variance (sigma^2), that is
the variance in the response variable conditional on all the fixed
and random effects. How do I calculate, or convince Stata to produce,
the variance in the response conditional only on the clustering
(sigma_0^2)?

Xu's formula for Omega^2 as a measure of the explained variation is
the ratio of the variance of the response conditional on all fixed
and random effects (sigma^2) and the variance of the response
conditional on the clustering only (sigma_0^2) subtracted from 1. It
is the proportion of the variance conditional on the clustering that
is explained by the covariates. That is:

Omega^2 = 1 - (sigma^2 / sigma_0^2)

This seems to be simple partitioning of the variance that I believe
SAS gives you automatically, but I'm not sure how to get this from
Stata.

Can anyone help me here?

______________________________

Patrick Phillips Statistician MRC Clinical Trials Unit 222 Euston

Email: ppp@ctu.mrc.ac.uk Tel: 0207 670 4637 Fax: 0207 670 4829

*********************************************************************************

This communication is confidential and may contain privileged
information intended solely for the named recipient(s). It may not be
used or disclosed except for the purpose for which it has been sent.
If you are not the intended recipient, you must not copy, distribute,
take any action or reliance on it. If you have received this
communication in error, do not open any attachments but please notify
the MRC Euston Campus ICT Service Desk by e-mailing ts@ctu.mrc.ac.uk
quoting the sender details, and then delete this message along with
any attached files. E-mail messages are not secure and attachments
could contain software viruses which may damage your computer system.
Whilst every reasonable precaution has been taken to minimise this
risk, The MRC cannot accept any liability for any damage sustained as
a result of these factors. You are advised to carry out your own
virus checks before opening any attachments. Unless expressly stated,
opinions in this message are those of the! e-mail author and not of
the Medical Research Council (MRC).

*********************************************************************************

* *   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/

*
*   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/