[R-sig-ME] using MuMIn with glmmADMB

Rafael Mares raffmares at gmail.com
Fri Apr 20 14:16:56 CEST 2012


Dear all

Has anyone tried to use MuMIn along with glmmADMB? I'm trying to use
the importance() function from the MuMIn package (version 1.7.7) to
the get the sum of AIC weights across models that include a given
explanatory variable, within a set of candidate models fitted using
glmmADMB (version 0.7).

I've used this for models fitted with lme4, and it works like this:
importance(c(mod0, mod1, mod2, mod3))
where mod[i] is the model name.

This doesn't seem to work for models fitted with glmmADMB, as I get:
"Error in formula.default(x) : invalid formula"

I would be grateful if anyone had ideas/suggestions on how to work
around this problem. I can create a table with the AIC weights for
each model without using MuMIn, but importance() quickly does the
final step of adding these weights for all of my explanatory
variables.
Thank you very much.

Raff



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