[R-sig-ME] AIC(c) tables from a set of glmer (lme4) models

[Ben Bolker]

Adriaan De Jong <Adriaan.de.Jong at ...> writes:

> 
> Hi,
> Is there a way to create AICc tables from a set of glmer
> poisson models created under lme4? The AICcmodavg
> package description doesn't include references to glmer. 
> Or is the AICc.mer function applicable on glmer's?
> Thanks in advance for any suggestions/hints.
> Cheers,
> Adjan


  I can't see why AICc.mer shouldn't be applicable to
GLMM fits (except for the fundamental objection that AICc
was developed in the context of linear models and might
not actually be appropriate for GLM[M]s at all -- see Richards
(2005) Ecology "Testing ecological theory using the information-theoretic
approach" for some relevant simulation results -- but this is a
theoretical issue, not one that can be fixed by using different
software ...)

library(lme4.0)  ## I'm using lme4.0, should be equiv to lme4 from CRAN
library(AICcmodavg)
example(glmer)
AICc(gm1)

seems to work fine.