Hi, I work on a zero-inflated count model (zero-inflated Poisson and zero-inflated negative binomial) to describe the number of detected H5N1 outbreaks in the subdistricts of Thailand. I would like to test the residuals of my model for spatial autocorrelation. But I don't really know how to test it as my model is not a conventional linear model nor a glm (rather a hierarchical model). Does a function (such as moran.test() or lm.morantest()) exist for residuals of hierarchical models (zero-inflated models) ? I cannot find any clue in the litterature... Would you mind to light me ? If nothing exists yet, what do you advise me to do ? Thank you very much for you answer. Best regards, Timoth?e
Timoth?e Vergne, DVM, MSc PhD candidate e-mail : timothee.vergne at cirad.fr CIRAD - UR AGIRs TA C22/E Campus international de Baillarguet 34398 Montpellier FRANCE tel : +33 4 67 59 38 68 ANSES - Laboratoire de Sant? Animale de Maisons-Alfort 23 avenue du G?n?ral de Gaulle 94706 Maisons-Alfort tel : +33 1 49 77 26 75