[R-sig-ME] Trouble running full ZINB or ZANB model: NaNs produced and errors in optim()

[Ben Bolker]

J Straka <jstraka at ...> writes:

> 
> Hello,
> 
> I’m attempting to run zero-inflated and hurdle models using the pscl()
> package, but keep encountering problems with running my
> fully-parameterized model and I’m not sure why.
> My data consist of measurements of seed set (the response, which is
> zero-inflated and overdispersed) for two treatments and a control (B,
> F, and O) at three elevations (low, mid, and high).  At each elevation
> I also measured average temperature (AvgTemp) and abundance of insects
> (N).  My goal is to construct a model that best describes seed set
> using the four predictors above (and interactions, if necessary).
> I find that I have no problems running a partial model with only my
> two categorical predictors as follows (just an example using hurdle(),
> but results are the same for zeroinfl()):

  [snip snip]

 This isn't a mixed model question, so it really belongs somewhere
else (r-help or r-sig-ecology spring to mind).

   I wouldn't be shocked if it turned out that your data set was
too small and/or noisy to support what you are trying to do (rule
of thumb is that you probably need about 10 observations per parameter
you are trying to estimate).  I would also suggest centering
(and possibly scaling) your continuous predictors (see
e.g. Schielzeth 2010 Methods in Ecology and Evolution)