Ian
The use of regression slope to measure kriging
efficiency or confidence in an estimate is spurious.
The slope of the regression line has nothing to do
with the correlation between true value and estimated
value and everything to do with whether or not the
standard deviations of these two variables are
comparable. In effect, it is a measure of conditional
bias, not a measure of correlation.
In fact, if you are using a least squares regression
on Normalised data, with the same units on both
variables, the standard deviation of the estimated
values would have to be lower than that of the true
values to obtain a slope as high as 0.9 -- because
least squares is minimising the vertical distance to
the line, not the 'true' distance to the line.
Better to use the kriging variance for classification
needs.
For example, you could insist that estimates for
(large) blocks of ground lie within a certain
confidence percentage of the true value. Companies
liek Anglo American use a criteria that the average of
the first year's production must be with 15% of the
true value at 90% confidence, before they will call it
"measured".
Others companies (such as Iskor) use what we call the
ygiagam criterion:
Measured: where kriging variance is less than original
sample variance (total sill) less within block
variance. That is, an estimate can be placed on teh
block with more confidence than simply allocating a
regional average.
Indicated: within the range of influence of at least
four boreholes.
Inferred: anything the geologist thinks is there.
Hope this helps.
Isobel Clark
http://uk.geocities.com/drisobelclark
____________________________________________________________
Do You Yahoo!?
Get your free @yahoo.co.uk address at http://mail.yahoo.co.uk
or your free @yahoo.ie address at http://mail.yahoo.ie
--
* To post a message to the list, send it to [EMAIL PROTECTED]
* 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 [EMAIL PROTECTED] 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
