[R-sig-ME] Estimating repeatability from ZIP models (MCMCglmm)

[Ned Dochtermann]

Hello,

I am working with some data and attempting to calculate repeatabilities for
individuals using ZIP models in MCMCglmm and was hoping for a bit of
clarification. 

I had previously calculated repeatabilities for Poisson models based on the
additive equations presented in Nakagawa and Schielzeth (2010, Biol. Rev)
using subject intercepts, unit residuals and the fixed effect intercept.
However, I have reanalyzed the data using a ZIP model due to substantial
zero-inflation.

I understand that the fixed coefficients are interpreted in the same manner
as they would be with a regular poisson model (with the exception of the
second latent) but I haven't seen anything discussing the effects of a ZIP
model on the random structure and thus repeatability.

Is it appropriate to calculate repeatability using the same variance
components and the same formula for a ZIP model as for a Poisson model?


Thanks for any advice,
Ned

--
Ned Dochtermann
Department of Biology
University of Nevada, Reno

ned.dochtermann at gmail.com
http://wolfweb.unr.edu/homepage/mpeacock/Dochter/
--