Re: AI-GEOSTATS: large error correlation coefficient?
> I have performed a jacknife estimation of e variableSorry, that's not true. The error is independent of
> V1* in a 3D space, the reference file have the true
> values of V1. Then I get the local error as
> ERR=V1-V1*. The problem is that correlation
> coefficient have large values for V1 and the other
> variables (0.7-0.9) and by theory the error must to
> be independent of the V1.
V1* but never of V1, especially if you are using a
linear estimator such as kriging.
> Why it happen, it is really dangerous in resourceThis is known as conditional bias and, in short, YES!
This effect has been documented for 50 years and has
variously been known as the regression effect, the
mine call factor, the volume/variance effect,
conditional bias and many other things (not all of
Want to chat instantly with your online friends? Get the FREE Yahoo!
* 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