[R-sig-ME] glmmadmb

[Kevin Burls]

Hi all-
I am comparing dispersal distributions from an artificial selection experiment on movement that contains replicated lines within two treatments over many generations. I'm actually not measuring movement in this particular case, but the placement of eggs by the organism as it moved within a generation. Thus, the (discrete) response data are the distance of eggs from a starting location. The dataset is fairly large (207,753 rows).  The data appear approximately poisson-distributed, and I am concerned about overdispersion as I am I was specifically selecting on the tail of the curve. I have already done a glmm with a poisson distribution but would like to try using the negative binomial. I have read glmmadmb is the way to model this but I run into the following error:

> dist.glmer<-glmmadmb(distance~treatment*generation,data=fullexpansion,family="nbinom",link="log",random=~1|line)
Need to increase the maximum number of separable calls allowed to at least 20001
Current value is 20000
Use the -ndi N command line option
Error in glmmadmb(distance ~ treatment * generation, data = fullexpansion,  : 
  The function maximizer failed (couldn't find STD file) Troubleshooting steps include (1) run with 'save.dir' set and inspect output files; (2) change run parameters: see '?admbControl'
In addition: Warning message:
running command './glmmadmb -maxfn 500 -maxph 5 -noinit -shess' had status 1 

This seems to be a size issue but perhaps it is something else? Is there a workaround for this or should I just attempt another method like quasilikelihood estimation?

Thanks for your help-
Kevin Burls


Kevin Burls
Ph.D. candidate
EECB Program
University of Nevada, Reno
kburls at unr.edu
http://wolfweb.unr.edu/~kburls