[R-sig-Geo] question about autoKrige anis1 and anis2

[Jon Skoien]

The result you got is the same as for a variogram model from vgm with a 
nugget effect. This is the ratio of anisotropy, which is one both for 
the nugget and for the correlated part of the model. See:

str(vgm(10, "Exp", 300))
str(vgm(10, "Exp", 300, 3))

I guess you see something like the second one as autofitVariogram is not 
able to fit the anisotropy parameters. For that you could have a look at 
estimateParameters in the intamap package, which can fit a variogram 
model of the vgm-type with anisotropy parameters.

Cheers,
Jon


On 7/29/2014 6:56 PM, Hodgess, Erin wrote:
> Hello again!
>
>
>
> I was using autoKrige (finally successfully!) and was looking at the model results.  I got anis1 of 1,1 and anis2 of 1,1.
>
>
>
> How does this relate to the anis in the vgm function, please?
>
>
>
> Thanks,
>
> Erin
>
>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo