[R-sig-Geo] predict.spautolm

[Roger Bivand]

On Mon, 21 Sep 2009, Bjarke Christensen wrote:

>
> I am looking for a function to make out of sample predictions using
> spautolm. The question has been asked before, two years ago, but with no
> response: http://finzi.psych.upenn.edu/R/Rhelp02/archive/113162.html.
> Certainly, someone must have needed one since?

Clearly not. Predict using just the regression coefficients (with 
non-spatial weights if used for fitting). If you want simulations around 
those point estimates, you can introduce spatial autocorrelation with the 
lambda term, but lambda will not impact the predictions unless the model 
was misspecified, I believe.

>
> Alternatively, does anyone have a reference or any hints wrt making such a
> function? Is there perhaps some reason not to do it?
>

It could be done, but isn't a high priority - unless you are using 
non-spatial weights, the speculative predict() method for sarlm objects 
from errorsarlm() will manage OK.

Hope this helps,

Roger

> Thanks in advance,
> Bjarke Christensen
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, 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