A linear model for coregionalisation for CoKriging

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

A linear model for coregionalisation for CoKriging

yong.t.li
Dear Edzer,

I have a question for the LMC in using cokriging.
I calculate the direct and cross semivariograms of a target variable and a co-variable and try to fit a LMC to these variograms. With gstat, I am suggested to obtain a LMC with the same range, but with different sills. However, in my case, the models have to be with different ranges and sills as well. I fit them all, and work. I use the LMC with different ranges and sills to do the interpolation and get a surface which is considered better than OK and RK as well. Does it make any sense to you because I use different ranges in my excercise?

By the way, when I read the documentation of current ILWIS, its cokriging function applies the LMC with different ranges and sills. I wish I read the right part of doc.

Regards,

Yong
_______________________________________________
Geostatistics mailing list
[hidden email]
http://list.52north.org/mailman/listinfo/geostatistics
http://geostatistics.forum.52north.org
Reply | Threaded
Open this post in threaded view
|

Re: A linear model for coregionalisation for CoKriging

Edzer Pebesma
Administrator


On 04/18/2011 06:13 AM, [hidden email] wrote:
> Dear Edzer,
>
> I have a question for the LMC in using cokriging.
> I calculate the direct and cross semivariograms of a target variable and a co-variable and try to fit a LMC to these variograms. With gstat, I am suggested to obtain a LMC with the same range, but with different sills. However, in my case, the models have to be with different ranges and sills as well. I fit them all, and work. I use the LMC with different ranges and sills to do the interpolation and get a surface which is considered better than OK and RK as well. Does it make any sense to you because I use different ranges in my excercise?

What can I say? Be happy when your kriging variances are not negative,
and hope you work with a consistent model!

>
> By the way, when I read the documentation of current ILWIS, its cokriging function applies the LMC with different ranges and sills. I wish I read the right part of doc.

a.f.a.i.k., that can't be.

With best regards,

>
> Regards,
>
> Yong
> _______________________________________________
> Geostatistics mailing list
> [hidden email]
> http://list.52north.org/mailman/listinfo/geostatistics
> http://geostatistics.forum.52north.org

--
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      [hidden email]
_______________________________________________
Geostatistics mailing list
[hidden email]
http://list.52north.org/mailman/listinfo/geostatistics
http://geostatistics.forum.52north.org