[R-sig-Geo] factorial kriging, or kriging separate spatial scales based on nested variogram

Edzer Pebesma edzer.pebesma at uni-muenster.de
Wed Dec 19 22:40:15 CET 2007


Felipe G. Nievinski wrote:
> Hi.
>
> Is factorial kriging supported in any of the R packages, please?
>
> I see how to model a nested variogram, but I don't see how I could krige 
> (predict) each component separately. It doesn't seem much more 
> complicated than the supression of the nugget effect, though.
>   
It is a bit more than that; I think it only really makes sense if you 
have two or more variogram components besides the nugget. It is not so 
hard to do, but I'm afraid none of the package developers has felt the 
urge to do so. Even nested variograms, e.g. with two spherical models, 
are afaik only found in one package, but I may be wrong there.
--
Edzer
> For an example, see slides 33, 34 in [1]; more details on pages 342-349 
> of Chiles & Delfiner.
>
> Margaret A. Oliver
> Making Sense of Sensed Data Using Geostatistics
> <http://users.ox.ac.uk/~uauug/011105conf/Margaret.ppt>
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>




More information about the R-sig-Geo mailing list