Message-ID: <498BE101.1070408@uni-bayreuth.de>
Date: 2009-02-06T07:04:33Z
From: Christina Bogner
Subject: inference for random effects
> Message: 5
> Date: Thu, 5 Feb 2009 13:48:08 -0500
> From: "Jeff Evans" <evansj18 at msu.edu>
> Subject: [R-sig-ME] inference for random effects
> To: <r-sig-mixed-models at r-project.org>
> Message-ID: <392F001D52E34A10868571F30AB625DD at myelin>
> Content-Type: text/plain; charset="us-ascii"
>
> I'm sure this must have been discussed before, but in searching the archives
> I haven't found an answer yet.
>
> Simple question:
>
> In lme4 can I evaluate the significance of a random effect in a model by
> substituting an uninformative dummy variable for it and comparing it to the
> model with the "real" random effect using anova?
>
> M1 = lmer(cbind(successes, total-successes) ~ A * B + (1|C), data=dat,
> family="binomial")
>
> M2 = lmer(cbind(successes, total-successes) ~ A * B + (1|Cdummy) , data=dat,
> family="binomial")
>
> anova(M1,M2)
>
> Where A, B, and C are factors, and Cdummy is a column with the word "dummy"
> in every row.
>
> Then compare the AIC, subtracting 2 from the M2 AIC score since it "falsely"
> estimated a parameter for the random effect. When I do this, I get delta AIC
> of about 600 favoring the more informative M1. Is this approach
> fundamentally wrong?
>
>
> Thanks,
>
> Jeff Evans
> Michigan State University
Hello Jeff,
for lme and lmer models there is a simulation package RLRsim by Fabian
Scheipl "for testing the presence of variance components" that might be
helpful.
Christina