[R-sig-ME] Observation-level random effect in poisson GLMM

[Jarrod Hadfield]

Hi,

Breslow refers to this approach as the "classical" approach in 1981 so  
I would imagine the idea is so old it could be attributed to common  
sense.

He cites HINDE J GLIM 82-109 1982 as using a log-normal mixing  
distribution instead of the usual gamma (i.e. negative binomial) which  
is equivalent to the observation-level random effect with log link.

Cheers,

Jarrod




Quoting Tim Carnus <Tim.Carnus at ucd.ie>:

> Dear all,
>
> I am looking for a reference(s) supporting the use of  
> observation-level random effects to deal with overdispersion in  
> poisson generalised linear mixed models.
>
> Best regards,
>
> Tim Carnus
>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>



-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.