[R-sig-Geo] Testing for spatial autocorrelation in residuals from a ZI model

[Timothee Vergne]

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,
Timothée

-- 
Timothée 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 Santé Animale de Maisons-Alfort
23 avenue du Général de Gaulle
94706 Maisons-Alfort
tel : +33 1 49 77 26 75