Message-ID: <4D3D3F9B.3090201@gmail.com>
Date: 2011-01-24T09:00:11Z
From: David Jansen
Subject: Question on overdispersion
In-Reply-To: <20101119095200.23134d5s4a1whrcw@www.staffmail.ed.ac.uk>
Dear Jarrod,
recently you suggested that to overcome overdispersion in a binomial distribution by fitting a observation-level random effect.
I was wondering if this method has been described anywhere so I can put a reference in my report.
Thanks a lot,
Greetings David Jansen
Hi Thierry + nameless,
It is not necessary to expand the binomial into Bernoulli trials (nor
advisable if n and/or the binomial size are large). You can just fit
observation-level random effects:
dataset$resid<-as.factor(1:dim(dataset)[1])
fit3<- glmer(cbind(male_chick_no, female_chick_no) ~ 1+(1|FemaleID)+
(1|resid), data = dataset, family = binomial)
gives the same answer as fit2
Cheers,
Jarrod
--
----------------------------------------------
David Jansen
PhD student in Vocal communication in banded mongoose