[ADMB Users] Problems using glmm.admb in R

[H. Skaug]

It seems that your data are fairly challengning in some way.
Try the argument

easyFlag = FALSE

glmm.admb is fairly robust, but there is no guaranty that it
works on every dataset. However, we have enough experience
to believe it works in general.

Hans

On Tue, Aug 17, 2010 at 1:17 AM, Richard Feldman
<richard.feldman at mail.mcgill.ca> wrote:
> Hello,
>
> I have been attempting to model several datasets in R using glmm.admb and
> have come across the following problem.
>
> The model is:
>
> model<-glmm.admb(Y ~ X1 + X2, random = X1, group = "SITE",
>
> with one of these further specifications:
>
> family = "poisson", zeroInflation = FALSE
> family = "poisson", zeroInflation = TRUE
> family = "nbinom", zeroInflation = FALSE
> family = "nbinom", zeroInflation = TRUE
>
> With the first three specifications, the function maximizer fails (or the
> model fitting has taken so long, I just give up). The modeling works fine
> for the last specification.
>
> I have tried this with several datasets, some zero-inflated and some not. I
> have the same problem every time: only "nbinom" and "zeroInflation=TRUE"
> works.
>
> I am running R 2.10.1 on Windows Vista x32.
>
> Has anyone else encountered this problem?
>
> Thank you,
>
> Richard
>
> --
> Richard Feldman, PhD Candidate
> Dept. of Biological Sciences, McGill University
> W3/5 Stewart Biology Building
> 1205 Docteur Penfield
> Montreal, QC H3A 1B1
> 514-212-3466
> richard.feldman at mail.mcgill.ca
> _______________________________________________
> Users mailing list
> Users at admb-project.org
> http://lists.admb-project.org/mailman/listinfo/users
>