RE: Cathy Schloeder's question

First of all one has to ask what is meant by the

lack of spatial autocorrelation and essentially this

has to be asked in the context of a model. Without

spatial correlation then the best interpolator is

the mean and then the question is whether this is

constant or dependent on position.

A non-constant mean would correspond to the use

of a Trend Surface. Note that the non-constant

mean might be in reference to some other characteristic

than position coordinates.

Inverse Distance Weighting, Splines, Kriging, Radial

Basis Functions all assume a form of spatial

correlation.

Without spatial correlation and a constant mean

simply use the average of the data values at each

point to be interpolated. Even at data points use

the average since then the difference is simply

"error"

Donald E. Myers

Dept Math.

University of Arizona

htttp://www.u.arizona.edu/~donaldm

--

*To post a message to the list, send it to

ai-geostats@....

*As a general service to list users, please remember to post a summary

of any useful responses to your questions.

*To unsubscribe, send email to

majordomo@... with no subject and

"unsubscribe ai-geostats" in the message body.

DO NOT SEND Subscribe/Unsubscribe requests to the list!