<html>
<font size=3>Dear R-listers,<br>
I would like to carry out a very basic descriptive analysis of homicides
rates in Italy, taking into account both the spatial dimension (103
provinces) and the temporal dimension (10 years), but no covariates. In
practice, what I would like to do is to describe spatio-temporal
variation of homicide rates, identifying those combinations of
province-year where the homicide rate is either over- or underrepresented
based on Poisson expectation.<br>
Is it possible to carry out such an analysis with R? If so, what package
should be preferred? GLMMGibbs, geoRglm, and function glmmPQL in MASS
package sound as possible choices to me, but I'm no expert of this kind
of analysis and I'd much appreciate any indication. <br>
Thank you very much for any help and best regards,<br>
Maurizio Pisati </font><br>
<br>
<br>
<br>
<br>
<div><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><></div>
<div> Maurizio Pisati</div>
<div> Associate Professor</div>
<div> Department of Sociology and Social Research</div>
<div> University of Milano Bicocca </div>
<div> Edificio U7</div>
<div> Via Bicocca degli Arcimboldi, 9 </div>
<div> 20126 Milano MI</div>
<div> Italy</div>
<div> Tel: +39-02-64487533</div>
<div> Fax: +39-02-64487561</div>
<div> E-mail: maurizio.pisati@galactica.it</div>
<div><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><></div>
</html>