[R-sig-ME] Compare glmmQPL

Ben Bolker bolker at ufl.edu
Tue Feb 9 17:59:17 CET 2010


Donal Bisanzio wrote:
> Good morning,
> 
> I made two different model with glmmQPL function. I need to compare them,
> but anova() does not work. I also changed class from "glmmQPL" "lme" to
> "lme", but this trick does not work.
> 
> Is there a way to compare these type of model??
> 
> Thanks for your help
> 
> Donal
> 

  It's tricky.
  What is the relationship between your models?  What kind of comparison
do you want to make? If they are nested, you can look at the Wald t-test
(or F-test, if you need to set multiple parameters to zero
simultaneously) of the parameters that differ.  If they are not nested,
then you can in principle compute a quasi-AIC value to compare them, but
 it's a bit tricky to extract that information ...

  Ben Bolker

-- 
Ben Bolker
Associate professor, Biology Dep't, Univ. of Florida
bolker at ufl.edu / people.biology.ufl.edu/bolker
GPG key: people.biology.ufl.edu/bolker/benbolker-publickey.asc




More information about the R-sig-mixed-models mailing list