An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20130925/66902573/attachment.pl>
DIC of SAR model
3 messages · Milan Sharma, Pascal Oettli, Roger Bivand
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20130926/bb120987/attachment.pl>
On Thu, 26 Sep 2013, Pascal Oettli wrote:
Hello,
require(sos)
findFn('deviance information criteria')
But I don't know whether it is usable with "spautolm" output. You will have to give it a try.
The underlying question is whether the paper fitted the simultaneous autoregressive model (spatial error model, SEM), or the spatial lag model (a.k.a. SAR) using Bayesian methods. If it did, then reporting DIC seems possibly sensible, if it used ML, it doesn't make much sense anyway. All kinds of odd things appear in published work, it's worth thinking through whether they should be there. Roger
Regards, Pascal 2013/9/26 Milan Sharma <milansharma2012 at yahoo.com>
Hi all, I read in a paper that has calculated the DIC (deviance information criteria) value of SAR model. Using spautolm (), I fitted SAR model, and DIC() does give an error.
sar<-spautolm(Y~X1+X2,data=data,listw=listw) DIC(sar)
Error: could not find function "DIC"
How do I calculate DIC value of SAR model?
I would appreciate any suggestions.
Milan
[[alternative HTML version deleted]]
_______________________________________________ R-sig-Geo mailing list R-sig-Geo at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand Department of Economics, NHH Norwegian School of Economics, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: Roger.Bivand at nhh.no