[R-sig-Geo] vgm with a Cauchy function

[Edzer Pebesma]

Francois JOUVIE wrote:
> Hello,
>
> I again send this below request to the List, because I do not receive  
> an answer to it, since July 8th.
> Perhaps anyone can help me ?
> Thanks you.
>
>
> My question:
>
>   I'm exploring the possibilities of the gstat package in the  
> R_Library context, to make variograms and kriging in a problem of  
> electromagnetic field.
>
>   This is my question:
>
>   To fit my points_variograms, the  vgm routine displays a closed set  
> of default functions ("Sph", "Gau", "Bes", etc.).
>
>   I'd like try a different fitting model (a Cauchy function  
> precisely). Do anyone knows if it is possible (and how to do it) to  
> use an owner_defined function to fit a
>   particular variogram ?
>   
There's two ways to do that.
1. you can pass an arbitrary function by computing its values and 
passing it as a table, an example is found in ?vgm:
     vv=vgm(model = "Tab",  covtable =
             variogramLine(vgm(1, "Sph", 1), 1, n=1e4, min = 0, 
covariance = TRUE))
2. you can edit the underlying C source code; if you can express the 
variogram function in terms of a C (or R) function, this is not as hard 
as it seems -- the code has been set up for easy addition of models.

You can also start working with packages that already provide this.
> Thanks you very much.
>
> (I'm new on this list and I don't know how to get the archived past  
> questions and answers)
>   
Just click on the link below any list message, and it will bring you 
there. Google searching with r-sig-geo as one of the key words should 
work as well.
--
Edzer
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> 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/