[ADMB Users] FW: glmm.admb in R

[H. Skaug]

Hi Zhanpan,

This is an error message from the ADMB program underlying glmm.admb().
I am not able to help you with this at the moment (maybe somebody else are).
We will look into this in future versions of the package.

Sorry for not being able to help you out.

Hans

On Fri, Feb 11, 2011 at 12:00 PM, Hans Julius Skaug
<Hans.Skaug at math.uib.no> wrote:
>
> ________________________________
> From: Zhanpan Zhang [mailto:zhanpan.zhang at email.ucr.edu]
> Sent: Thursday, February 10, 2011 10:41 PM
> To: users at admb-project.org; Hans Julius Skaug
> Subject: glmm.admb in R
>
> Dear Professor Skaug,
> I used glmm.admb() to analyze a dataset, and R keeps displaying the error
> message "Error matrix not positive definite in choleski_decomp" and it seems
> that the program keeps running and won't stop. Could you please help check
> this? Thank you very much.
> Attached is my dataset with two columns of "y" being the response and
> "subject" being the random group factor, and here is my code:
> library(glmmADMB)
> data=read.table("data.txt",header=TRUE)
> problem=data.class(try(glmm.admb(y~1,random=~1,group="subject",data=data,family="nbinom"),silent=T))=="try-error"
>
> Sincerely,
> Zhanpan
>
>