Thank you, please forgive my naivety but I have tried the krigeTg and I
couldnt add any cokriging formula into it (or gstat object). it seems that
it only accepts the univariate formula and we cant use it for cokriging.
like this:

krigeTg(V1 ~ V2, locations,newdata......


On Fri, Mar 19, 2010 at 4:45 AM, Edzer Pebesma <
[email protected]> wrote:

> What happens in kigeTg doesn't look that hard. Have you tried?
>
> Ebrahim Jahanshiri wrote:
> > Thank you very much.
> > However I found no couterpart for that in cokriging. Is there any way to
> do
> > that manually? I mean to get "m" manually?
> >
> > On Mon, Mar 1, 2010 at 8:26 PM, Edzer Pebesma <
> [email protected]
> >> wrote:
> >
> >> Ebrahim Jahanshiri wrote:
> >>
> >>> I am wondering if there is any function or option to back-transform the
> >>> result of the lognormal kriging in gstat yet.
> >>>
> >>>
> >> You can use krigeTg, which does trans-Gaussian kriging for all Box-Cox
> >> transforms; passing it a lambda parameter with value 0, log-normal
> kriging
> >> is used. Pls look at the examples, the back-transformed values have
> funny
> >> names like var1TG.pred.
> >>
> >> The whole think has been implemented for ordinary kriging only; if
> someone
> >> can point me out a source that describes how to do this for universal /
> >> external drift kriging, I'd be happy to look into how to implement it.
> >>
> >>> Although for simple kriging it is fairly straight forward, that is, it
> is
> >>>
> >>> possible to do back-transform using the vector of predicted values and
> >>> kriging variances, but for ordinary kriging, it might be difficult
> because
> >>> we dont have lagrange multiplier (according to my web help search) to
> do
> >>> the
> >>> back-transformation. I am wondering if there has been any advancement
> in
> >>> that matter?
> >>>
> >>> thank you
> >>>
> >>>        [[alternative HTML version deleted]]
> >>>
> >>> _______________________________________________
> >>> R-sig-Geo mailing list
> >>> [email protected]
> >>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
> >>>
> >>>
> >> --
> >> Edzer Pebesma
> >> Institute for Geoinformatics (ifgi), University of Münster Weseler
> Straße
> >> 253, 48151 Münster, Germany. Phone: +49 251 8333081, Fax: +49 251
> 8339763
> >> http://ifgi.uni-muenster.de http://www.52north.org/geostatistics
> >> [email protected]
> >>
> >>
> >
>
> --
> Edzer Pebesma
> Institute for Geoinformatics (ifgi), University of Münster
> Weseler Straße 253, 48151 Münster, Germany. Phone: +49 251
> 8333081, Fax: +49 251 8339763  http://ifgi.uni-muenster.de
> http://www.52north.org/geostatistics      [email protected]
>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

Reply via email to