Skip to content
Prev 950 / 7420 Next

low predicted vales in GAMs

Hi Anna,
A couple of thoughts:
Did you try fitting a straight Poisson model? The quasi Poisson model is
assuming
the variance is not a strict function of the mean, and that may be
interacting with your
weighting function. Also how exactly are the weights defined? is it (#
squares counted)/(total possible squares) ?
Or can weights be greater than 1?
Did you try fitting without weights and offset? 
And lastly are your bird counts highly clustered? ie lots of 0's and
then some high counts?
because the model will probably over smooth the high counts. 
Since this is a spatial model you might want to look at geoRglm. Or try
cozigam
and try to model zero-inflation (if that is the case)

Hope this helps
Nicholas