[R-sig-Geo] how to change the fit.method from autoKrige, please

[Hodgess, Erin]

Great....thanks so much!

Sincerely,
Erin

________________________________________
From: Jon Skoien [jon.skoien at jrc.ec.europa.eu]
Sent: Tuesday, July 29, 2014 2:06 AM
To: Hodgess, Erin; r-sig-geo at r-project.org
Subject: Re: [R-sig-Geo] how to change the fit.method from autoKrige, please

It seems you cannot change the fit.method from autoKrige or
autofitVariogram, it uses the default from gstat::fit.variogram, which
is method 7.
However, you should probably first try to see why you get the error and
what is zero. I assume that your error is that you have zero distances
semivariances, do you have a lot of duplicated locations? Is this
necessary for your case to keep these, can you delete them, or can you
add some noise to the coordinates with jitter?

Best wishes,
Jon


On 7/28/2014 10:22 PM, Hodgess, Erin wrote:
> Hello!
>
>
>
> I'm using the autoKrige function on a data set and keep getting an error that fit.method 7 has zero semivariances.
>
>
>
> How do I change the fit.method from within autoKrige, please?
>
>
>
> I've tried using it both inside and outside of miscFitOptions, but no luck yet.
>
>
>
> Any suggestions would be much appreciated.
>
>
>
> Sincerely,
>
> 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