AI-GEOSTATS: Re: non-ergotic covariance
- Thanks for the helpful responses.
I opted to use this estimator for model parameters because after
detrending my data, the residuals gave almost an identical
semivariogram to that of the sample data. So I concluded that
universal kriging was probably not the best way to go.
I checked out Ed Isaaks page about the correlogram and Issaks and
Srivastava for the non-erg cov. formula. I would prob. use a
correlogram except that I'm dealing with that pesky Variowin floating