I'm interested in running a zero-inflated Poisson/negative binomial model. Unfortunately, I have very little experience with these models and have not been able to find much online or in textbooks. Could anyone point me in the direction of resources that would help? Or would anyone be willing to walk me through the process? Interested in how to select appropriate estimation and validation methods as well as how to know I'm getting a good fit.
I am using JMP 13.
When I do a Google Search for "overdispersion model", I get about 339,000 results.
In the body of that search, there are many great resources: several short courses, the classic McCullagh and Nelder Generalized Linear Models book, the Morel and Neerchal Overdispersion Models in SAS book (this is really good!), github pdf's, all sorts of stuff.
Check 'em out. Some of the books can even be downloaded from those links.
Thanks for the reply.
Searching for overdispersion models rather than negative binomial GLM's has definitely produced more relevant results.
I've been skimming through the Generalized Linear Models book by McCullagh and Nelder and while it is undoubtedly incredibly informative, it's also a bit dense. At least for me. I'm a non-statistics graduate student and while I would love to gain a more in-depth understanding of GLM's, I don't necessarily have the time to run through a 500+ page book on an analysis that is a fairly small part of my thesis. If I'm being completely honest, I was looking for a more approachable and easily digestable resource that would walk me through the process within JMP rather than the mathmatical theory behind a GLM.
I realize that something like this is probably wishful thinking but just thought I'd ask.
Yeah, that McCullagh and Nelder book has the answers, but it is dense as a neutron star! :-)
Perhaps your school has a statistics department with whom you could consult?