[R-sig-Geo] help for xvalid with external data

[Paulo Justiniano Ribeiro Jr]

Dear Giovanni

Apparently there is a bug passing the trend
in the current version of the function.
I'll fix and let you know

Thanks
P.J.


Paulo Justiniano Ribeiro Jr
LEG (Laboratorio de Estatistica e Geoinformacao)
Universidade Federal do Parana
Caixa Postal 19.081
CEP 81.531-990
Curitiba, PR  -  Brasil
Tel: (+55) 41 3361 3573
Fax: (+55) 41 3361 3141
e-mail: paulojus AT  ufpr  br
http://www.leg.ufpr.br/~paulojus



On Wed, 9 Jul 2008, Giovanni Bacaro wrote:

>
>
> Hi, as usual I have a problem with R functions!
>
> I formulated a geostatistical model with geoR.
>
> I decided to include some covariates in my model, and I estimated all the
> parameters (linear model and covariance function parameters) by the likfit
> function as follows:
>
>
>
> mod1<-likfit(geodata, ini=varfitmodel, trend=~cov1 +cov2, lambda=0.5,
> cov.model="matern",kappa=0.5)
>
>
>
> Once I obtained my best model, I'd like to validate it using an external
> dataset by using the function xvalid.
>
>
>
> First of all, I transformed the external dataset in a geodata object, as
> usual, named
>
> "validation_data"
>
> ,where also covariates were included.
>
> Then I tried the following syntax:
>
>
>
> Validation<-xvalid(geodata, model=mod1,
> locations.xvalid=validation_data$coords,  data.xvalid= validation_data$data)
>
>
>
> Bur R says:
>
> Trend elements not found.
>
>
>
> My question is: how can I specify the trend structure within the xvalid
> function?
>
> Thank a lot
>
>
> 	[[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
>