Ali
Your question seems to be simple but could run into a very tight
philosophical type debate if one wants to.
You may want to do either of the following:
1. Read a chapter entitled "Search Strategy" in Isaaks' book and start
from there.
2. Identify the radius of influence (i.e., Range) using variogram modeling
and limit your search to all those data inside radius of influence.
Conceptualy, data outside radius of influence won't contribute to variance
reduction task. However, you may run into difficulty identifying radius of
influence as this measure of distance is scale dependent. Please be
advised that process range is very different from observation range (i.e.,
range obtained from data).
3. Use clustering algorithm to cluster your data (e.g., k-mean) and limit
your search to each cluster of data. Again, you may run into problem
deciding on the number of clusters to use (Very subjective).
Hope this helps
Thanks
Abedini
On Fri, 30 Jun 2006, Alí Santacruz wrote:
Dear list members,
I have a very simple question (I think):
When I want to perform a kriging, I must define the number of nearest
observations that should be used for the kriging prediction, or a maximum
distance from the prediction location.
What criteria should I use to set these parameters? Which is the optimum
number of nearest neighbors?
Any comment is welcome.
Sincerely,
Alí M. Santacruz
M.Sc. Geomatics, Student
National University of Colombia
Bogotá D.C.
_________________________________________________________________
Charla con tus amigos en línea mediante MSN Messenger:
http://messenger.latam.msn.com/
+
+ To post a message to the list, send it to [email protected]
+ To unsubscribe, send email to majordomo@ jrc.it with no subject and
"unsubscribe ai-geostats" in the message body. DO NOT SEND
Subscribe/Unsubscribe requests to the list
+ As a general service to list users, please remember to post a summary of
any useful responses to your questions.
+ Support to the forum can be found at http://www.ai-geostats.org/
