[R-sig-Geo] Assessing residual spatial autocorrelation in a Poisson

Roberto Patuelli roberto.patuelli at usi.ch
Fri Nov 27 01:25:17 CET 2009

Dear Karen,

I've also been working on spatial count (glm) models, in particular using 
spatial filtering (Griffith 2003), which helps you incorporating spatial 
autocorrelation in the mean response term.
Michael Tiefelsdorf has written a spatial filtering code in R, which is 
included in spdep. But it is based on a Moran's I minimization criterion 
only, and uses (if I'm not wrong) a linear model.
Instead I've used it on count regressions, by writing my own R code, which 
of course I'm willing to share :) For example, I will soon make the code 
used in this paper (http://ideas.repec.org/p/lug/wpaper/0806.html), which is 
forthcoming on the Annals of Regional Science, public.
See my website for more :))


Roberto Patuelli, Ph.D.
Istituto Ricerche Economiche (IRE) (Institute for Economic Research)
Università della Svizzera Italiana (University of Lugano)
via Maderno 24, CP 4361
CH-6904 Lugano
Phone: +41-(0)58-666-4166
Fax: +39-02-700419665
Email: roberto.patuelli at usi.ch
Homepage: http://www.people.lu.unisi.ch/patuellr

More information about the R-sig-Geo mailing list