R help - Jul 2008 - Question regarding lmer vs glmmPQL vs glmm.admb model on a negative binomial distributed dependent variable

Hi R-users,
 
I intend to apply a mixed model on a set of longitudinal data, with a negative
binomial distributed dependent variable, and after following the discussions on
R help list I saw that more experienced people recommended using lmer (from lme4
pack), glmmPQL (from MASS) or glmm.admb (from glmmADMB pack)  
 
My first problem: yesterday this syntax was ok, now I get this weird message (I
got it before when I was using my own set of data)
 
>        data(epil2)
>       
glmm.admb(y~Base*trt+Age+Visit,random=~Visit,group="subject",data=epil2,family="nbinom")
'C:/Documents' is not recognized as an internal or external command,
operable program or batch file.
Error in glmm.admb(y ~ Base * trt + Age + Visit, random = ~Visit, group =
"subject",  :
  The function maximizer failed
In addition: Warning messages:
1: In file.remove(std_file) :
  cannot remove file 'nbmm.std', reason 'No such file or
directory'
2: In shell(paste(.path.package("glmmADMB"), "/admb/",
file_name, ".exe",  :
  'C:/Documents and Settings/eugen.pircalabelu/My
Documents/R/win-library/2.7/glmmADMB/admb/nbmm.exe -maxfn 500 ' execution
failed with error code 1
 
Second problem: when running these two commands on the data set epil2 from
glmmADMB package, I get very different results (I see that lmer gives SE twice
as big as those returned glmmPQL) and I was wondering which algorithm is best
suited for a highly skewed dependent variable (running these two commands on my
data set the diffrence in SE estimation still remains, and I want to use a
correct algorithm for my data set) .
 
Thank you very much and have a great day ahead!   
 
Eugen Pircalabelu


      
	[[alternative HTML version deleted]]