[R] R2 in mixed model
Jean G. Orelien
jorelien at scimetrika.com
Mon Mar 20 16:11:02 CET 2006
You can also refer to the following papers which have proposed Rsquares for
the linear mixed model:
Vonesh, E. F., Chinchilli, V. P., and Pu, K. W. (1996), "Goodness-of-fit in
generalized nonlinear mixed-effects models," Biometrics, 52, 572-587.
Zheng, B. Y. (2000), "Summarizing the goodness of fit of generalized linear
models for longitudinal data," Statistics in Medicine, 19, 1265-1275.
Xu, R. H. (2003), "Measuring explained variation in linear mixed effects
models," Statistics in Medicine, 22, 3527-3541.
Vonesh, E. F. and Chinchilli, V. M. (1997), Linear and nonlinear models for
the analysis of repeated measurements, Marcel Dekker. (pages 419-427)
Sadly, I have found based on simulations that these pseudo-Rsquares don't
work very well. I will be presenting a paper at JSM in August where I'm
proposing a new statistic for linear mixed models where you have conditional
independence. The manuscript has not gone through the peer-review process
Jean G. Orelien
From: r-help-bounces at stat.math.ethz.ch
[mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Spencer Graves
Sent: Friday, March 17, 2006 11:14 PM
To: Angelo Colombo
Cc: r-help at stat.math.ethz.ch
Subject: Re: [R] R2 in mixed model
The coefficient of deterimination, R^2, is defined as the percent
the variance explained. For a mixed model, we need to ask, "percent of
WHICH variance?" For more comments, I tried "RSiteSearch" with various
kew words. The most relevant comment I found was
"http://finzi.psych.upenn.edu/R/Rhelp02a/archive/48023.html", which I
got from RSiteSearch("R squared with mixed effects").
hope this helps.
Angelo Colombo wrote:
> I have some basic question for you!.
> Has calculating R2 sense in mixed model? I think no! But i don't why!
> Thank in advance for your help
> R-help at stat.math.ethz.ch mailing list
> PLEASE do read the posting guide!
R-help at stat.math.ethz.ch mailing list
PLEASE do read the posting guide!
More information about the R-help