[R] negative binomial lmer

[Bill.Venables at csiro.au]

lmer will work with negative binomial models, provided you specify an
explicit, scalar value for theta.

Rather than family = negative.binomial(theta = x) try something like
family=negative.binomial(theta = 2.5) (or whatever you wish specify as
theta).

Bill Venables.
 

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On Behalf Of H. Skaug
Sent: Thursday, 15 November 2007 8:16 PM
To: r-help at r-project.org
Subject: [R] negative binomial lmer

If lmer() does not do it, you can try:

http://otter-rsch.com/admbre/examples/glmmadmb/glmmADMB.html


It handles negative binomial responce (but you may have to remove data
entries involving NA manually).

Regards,

hans

>Hi
>I am running an lmer which works fine with family=poisson
>
>mixed.model
<-lmer(nobees~spray+dist+flwabund+flwdiv+round+(1|field),family="poisson
",method="ML",
na.action=na.omit)
>
>But it is overdispersed. I tried using family=quasipoisson but get no P
>values. This didnt worry me too much as i think my data is closer to
>negative binomial but i cant find any examples of negative binomial
lmer. I
>tried using the family=negative.binomial(theta=x,link=log) but got an
error
>message from R saying the function famiily=negative.binomial wasnt
>recognised.
>
>Can anyone suggest how to go about setting up the lmer with negative
>binomial distribution?
>
>Thanks

	[[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.