Dear list,
I am trying to decide if I have intrinsic correlation in a multivariate
dataset. According to Wackernagel (2003) (p.156-157) there are two ways
to test for intrinsic correlation:
a) Calculate codispersion coefficients (i.e. the ratio of the cross
versus the direct variograms). If the codispersion coefficients are
constant, we can assume intrinsic correlation.
b) Calculate the cross-variograms on the principal components of the set
of variables. If the cross variograms are zero at all lags, then we have
intrinsic correlation.
When I apply the methods described above to my data, it is difficult to
decide if the codispersion coefficients have enough trend to indicate
that there is no intrinsic correlation. In the same way, it is difficult
to decide when cross-variograms on the principal components are
“sufficiently” different from zero. Is there a statistical test to
detect intrinsic correlation?
Julian M. Burgos
Fisheries Acoustics Research Lab
School of Aquatic and Fishery Science
University of Washington
1122 NE Boat Street
Seattle, WA 98105
Phone: 206-221-6864
+
+ 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/
