[R-sig-Geo] How to find GWR prediction variance as a measure of uncertainty?

[Roger Bivand]

On Sun, 15 Apr 2012, Muhammad Imran wrote:

>
> Hi list fellows,
>
> Do someone have any idea to calculate the standard deviation of GWR 
> predictions (prediction variances), similary as we calculate with the 
> sqrt of prediction variance provided by kriging?

As spgwr says when it loads:

"This package does not constitute approval of GWR as a method of spatial 
analysis"

The ability of GWR to fit local coefficients sensibly (given doubts about 
appropriate bandwidths) is not established. Consequently, making 
predictions is encumbered with more than just the uncertainty of fitting 
the local weighted regression. So it is not clear how you should proceed - 
using GWR is an exploratory technique at best until its properties have 
been checked.

Roger

>
> Would be thankful for any help?
>
> Best regards
>
> Imran
>
> _______________________________________________
> 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