Hi,
I would like to obtain the dispersion parameter for a quasipoisson model for
later use in calculating QAIC values for model comparison.Can anyone suggest a
method of how to go about doing this?
The idea I have now is that I could use the residual deviance divided by the
residual degrees of freedom to obtain the dispersion parameter. The residual
deviance is available in the summary output for the lmer model. The problem
with this idea is that the residual degrees of freedom is not directly
available. I have been assuming that the estimated/approximate DF can be
obtained through the formula listed below, and then subsequently used to obtain
the dispersion parameter as described above. Is this correct?
samp.m4 <- mcmcsamp(m4, n = 1000)
print(summary(samp.m4))
(eDF <- mean(samp.m4[,"deviance"]) - deviance(m4, REML=FALSE))
# potentially useful approximate DF?
However, rather than going through this roundabout, there appears to be an
easier way to obtain the dispersion parameter. I have noted that the
'quasibinomial' produces a dispersion parameter in the model output, but the
'quasipoisson' output does not contain this useful number. Given that each of
my models is fit by lmer with a quasipoisson distribution, the program must be
internally calculating the dispersion parameter as it runs. Perhaps there is a
way to obtain this number directly since I assume it has already been
calculated? Can somone who has experience with this code provide some advice on
whether this is possible and how I might manage to do this?
Thank you for your time,
Wayne Hallstrom
403-370-3832
---------------------------------
[[alternative HTML version deleted]]
______________________________________________
R-help@stat.math.ethz.ch 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.
