Dear All,
I am trying to model the habitat mapping of 4 sharks species and my data are
zero inflated and positively skewed. The zero percentages in my data vary
between 56-77%. I did some tests fitting GAMs with negative binomial family
(with theta between 1-10). Is this the best way to go? Some people have used
VGAM (with zero inflated negative binomial) but my concern is that using the
VGAM model I would lose the basic ecological assumption (e.g. if the habitat is
not suitable your species is not there = 0 abundance).
Thank you very much in advance for any suggestions!
Kind Regards,
Valentina
[[alternative HTML version deleted]]
