[R-sig-Geo] DIC of SAR model

[Roger Bivand]

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