[R] Fitting binomial lmer-model, high deviance and low logLik

[rahul143]

Hello 

I have a problem when fitting a mixed generalised linear model with the 
lmer-function in the Matrix package, version 0.98-7. I have a respons 
variable (sfox) that is 1 or 0, whether a roe deer fawn is killed or not 
by red fox. This is expected to be related to e.g. the density of red 
fox (roefoxratio) or other variables. In addition, we account for family 
effects by adding the mother (fam) of the fawns as random factor. I want 
to use AIC to select the best model (if no other model selection 
criterias are suggested). 

the syntax looks like this: 
 > mod <- lmer(sfox ~ roefoxratio + (1|fam), data=manu2, family=binomial) 

The output looks ok, except that the deviance is extremely high 
(1.798e+308). 

 > mod 
Generalized linear mixed model fit using PQL 
Formula: sfox ~ roefoxratio + (1 | fam) 
    Data: manu2 
  Family: binomial(logit link) 
            AIC           BIC         logLik      deviance 
  1.797693e+308 1.797693e+308 -8.988466e+307 1.797693e+308 
Random effects: 
      Groups        Name    Variance    Std.Dev. 
         fam (Intercept)      17.149      4.1412 
# of obs: 128, groups: fam, 58 

Estimated scale (compare to 1)  0.5940245 




-----
   TO GET MORE DETAILS CLICK HERE  





--
View this message in context: http://r.789695.n4.nabble.com/Fitting-binomial-lmer-model-high-deviance-and-low-logLik-tp4651669.html
Sent from the R help mailing list archive at Nabble.com.