Do some of your observations have weights=0? If yes, that might be
the problem. I don't know the details, but some functions like glm have
an argument 'subset' and might give inappropriate answers with weights =
0.
Beyond that, I just got 69 hits for RSiteSearch("glmmPQL weights").
Some of these might interest you. In particular, you might also try
fitting the same model using lmer associated with the lme4 package; for
that see the vignettes in the lme4 and MlmRev packages.
If you would like further help from this listserve, please provide a
simple, self-contained, reproducible example as suggested in the posting
guide! "www.R-project.org/posting-guide.html". Without it, I'm
just
guessing, and some who could help more than I can don't bother.
hope this helps,
spencer graves
Nelson, Kerrie wrote:
> Hello,
>
> I am using the R function glmmPQL to fit a logistic GLMM, with weights.
> I am finding that I get fairly different parameter estimates in glmmPQL
> from fitting the full dataset (with no "weight" statement) and an
> equivalent, shorter dataset with the weights statement. I am using the
> weights statement in the 'glmmPQL' function exactly as in the
'glm'
> function. I also tested my dataset in the glm function and successfully
> got the correct results using the full (with no weights) and the shorter
> dataset (with weights).
>
> Is the weight statement in glmmPQL intended to be applied in a different
> way to a dataset than how it is used in the glm function?
>
> Thank-you for your help and any suggestions,
>
> Kerrie
>
> +++++
> This mail has been sent through the MPI for Demographic Rese...{{dropped}}
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html