I am looking for some advice concerning techniques in R that are
appropriate for correlated count data.
Specifically, I have some "freezing days" data, which is a count of the
number of days each spring that were below freezing. The counts were taken
at the same location over a period of years. The data set is highly zero
inflated with a large amount of over-dispersion; glm with a quasipoisson
error structure would be appropriate, except that there is a high degree of
correlation at lags of 1 making something like a corAR1 structure
appropriate. My difficulty is that glm() does not take an argument for
correlation, and I have no random effects to make something like lmer a
useful alternative--at least I don't think it is, but I'm not terrible
experienced with mixed-effects models. Therefore I need some alternative to
look at trends in this data over time that combines the functionality of
gls() and glm().
My basic model is to look at trends in number of freezing days over time
and ideally looks like:
freezing days~year, family=quasipoisson, correlation=corAR1
Thanks,
Lee
