Jorg, that is correct.

The variance of any binary variable (such as an indicator variable, 0s and 1s) is, by definition, p*(1-p), where "p" is the percentile of the distribution.

Therefore, the maximum possible variance of an indicator variable will be 0.25, when p=0.5; also, the variance as a function of the percentiles is a symmetrical curve, where p=10 has the same variance as p=90.

The only comment is, that, strictly speaking and as you probably know, covariances and variograms are not always interchangeable.

Cheers,

Mario

"Dilling,_J�rg" <

Joerg.Dilling@...> wrote:

dear all,

i have performed indicator kriging with different thresholds (corresponding

to 10th percentiles). is it correct that the variogram of each

xth-percentile is bounded to:

max cov = xth_percentile* (1-xth_percentile) ??

e.g. max cov (30th perc.) = 0.21, max.cov (90th perc.) = 0.09

thanks for comments

regards

j�rg

**************************************************

Dipl. Geo�kologe J�rg Dilling

Bayerisches Geologisches Landesamt

- Au�enstelle Marktredwitz -

Leopoldstra�e 30

D - 95615 Markredwitz

Tel.: ++49 (0)9231 - 951 105

Fax.: ++49 (0)9231 - 951 102

**************************************************

--

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

ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.

* To unsubscribe, send an email to

majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list

* Support to the list is provided at

http://www.ai-geostats.org
---------------------------------

Do you Yahoo!?

SBC Yahoo! DSL - Now only $29.95 per month!

[Non-text portions of this message have been removed]