[R-sig-Geo] gstat warning message from cross validate linear model

[Edzer Pebesma]

You are using a variogram that is permissible in 1D, but not in 2D
(linear with sill). This will generate negative variances, from which
the square root (sqrt) is computed in order to produce z scores.

The advice is to not use this model for data spanning more than one
dimension; the software, or its author, assumed you know this.

On 04/07/2014 08:59 PM, Moshood Agba Bakare wrote:
> Hi,
> The four variograme models below are fitted to the same dataset. I got
> a *warning
> message *while cross validating a fitted linear variogram model as
> 
> 
> Warning message:In sqrt(ret[[var.name]]) : NaNs produced
> 
> 
> from this R script
> lin.kcv <- krige.cv(yield ~ 1, canmod.sp, model = lin.var,nmax=100, nfold =
> 5)
> 
> what is the meaning of this warning? Each time I run this R script, I got
> different values for the cross validation statistics of linear.
> 
> Model         RMSE    MSDR
> Spherical      0.66    1.61
> Gaussian       0.62    3.17
> Exponential    0.65    1.91
> Linear         95.91   83.25
> 
> 
> These valueslook very strange compared to others because of warning message
> above. How do I explain this?
>  I made effort to redefine my nfold to be 10 but still give the same
> warning.
> 
> Thanks
> Moshood
> 

-- 
Edzer Pebesma
Institute for Geoinformatics (ifgi), University of Münster
Heisenbergstraße 2, 48149 Münster, Germany. Phone: +49 251
83 33081 http://ifgi.uni-muenster.de GPG key ID 0xAC227795

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: OpenPGP digital signature
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20140407/80d9d0f4/attachment.bin>