[R-sig-ME] problem using lmer with family = Gamma

Rafael Mares crm53 at cam.ac.uk
Fri Jun 11 19:08:18 CEST 2010

Dear all

I'm trying to run a GLMM with Gamma errors (as seems most appropriate
for my data) using the lme4 package (version 0.999375-33) in R version
2.10.1 on Windows.

I measured the time an individual sniffed a sample (continous response
variable, no 0s and all positive), in an experiment with a control and
1 treatment, while also controlling for sex and dominance class and
the age of the donor of the sample. I have 2 random effects:
individual (n = 31) and group to which the individual belongs (n = 8).
There are no missing values.

The formula is:

fullmod<-lmer(sniff ~ treatment + sex + domstatus + donor.age +
(1|group/id), family = Gamma, REML = FALSE, data = pres)

and this is the output:

Error in mer_finalize(ans) : mu[i] must be positive: mu = -44.2351, i = 1

Does anyone have any idea on what might be going on?  Someone posted a
similar question about a month ago, but it doesn't seem like anyone
replied.  I'll happily send some of my data if someone is able to look
into this.
Thanks in advance for your help and my apologies if this is not the
right place to post this question.


Rafael Mares
Large Animal Research Group (LARG)
Department of Zoology
University of Cambridge
Downing Street

More information about the R-sig-mixed-models mailing list