[R-sig-Geo] Handling realisations in geoR

[Paulo Justiniano Ribeiro Jr]

Hi Anders

First of all you only use the argument realisation if you have
"repetions"of the process (which is not common in practice)

>From what you write it seems all oyu need is somathing along the lines:

obj.geo <- as.geodata(x)
obj.ml <- likfit(obj.geo, ini=c(?,?)

where x is a matrix with coordinates and data values
and ini requires initial values for the covariance parameters, for which
you can use the empirical variogram (function variog()) and an rough
estimate (function eyefit()) to obtain

Hope this helps, please have a look at
?likfit

and there is the geoR intro web page at:
http://www.est.ufpr.br/geoR

and a few tutorials at
http://www.est.ufpr.br/CE714

and feell free to contact me if you have further questions

best
P.J.




On Thu, 9 Dec 2004, Anders Malmberg wrote:

> Hi,
>
> I have replicates of a spatial field to which I would like fit a spatial
> covariance function
> using geoR and its function likfit. My problem is that I do not know how
> to specify this in geoR.
>
> I tried to simply rowstack my fields such that I have
>
>        x   y   data
> "1" x1 y1 d11
> "2" x2 y2 d12
> "3" x3 y3 d13
> "4" x4 y4 d14
> "1" x1 y1 d21
> "2" x2 y2 d22
> "3" etc...
> "4"
> "1"
> "2"
> "3"
> "4" x4 y4 d34
>
> stored in an matrix "obj" let's say. How shall I use
> as.geodata(obj,realisations=XXX)?
> And how should I use likfit??? It is no problem to rearrange my matrix.
>
> Best regards,
> Anders Malmberg
> andersm at maths.lth.se
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>
>

Paulo Justiniano Ribeiro Jr
LEG (Laboratório de Estatística e Geoinformação)
Departamento de Estatística
Universidade Federal do Paraná
Caixa Postal 19.081
CEP 81.531-990
Curitiba, PR  -  Brasil
Tel: (+55) 41 361 3573
Fax: (+55) 41 361 3141
e-mail: paulojus at est.ufpr.br
http://www.est.ufpr.br/~paulojus