Dear Dr Van den Boogaart, Dr Merks is my son, coauthor of Precision Estimates for Ore Reserves, and a lead architect behind the Eclipse Modeling Framework. We took the same stats courses at the same university but at different times and for different reasons. Heres is the same proposition once more: Each functionally dependent value has its own variance TRUE or FALSE? Ive asked you to look at Fig 203 on page 286 of Davids 1977 Geostatistical Ore Reserve Estimation. How many functionally dependent (calculated!) holes do you count? How many functionally independent (measured!) holes do you count?
http://www.geostatscam.com/degrees_of_freedom.htm http://www.geostatscam.com/counting_degrees_of_freedom.htm http://www.geostatscam.com/Adobe/1977%20David.pdf http://www.geostatscam.com/Adobe/CBul198903.pdf
It was not me who brought up the Cauchy algorithm and the KWBLUP method. All I want to know is who replaced the variance of a SINGLE distance-weighted average with the pseudo variance of a SET of degrees-of-freedom and variance-deprived functionally dependent distance-weighted averages. Even the most astute geostatistician could not possibly fake a non-zero intrinsic variance of bogus gold in a phantom resource. I do not need more prevarications about Bre-X, about whether to verify or not to verify spatial dependence, or about whether "the variance is an object defined for each random variable and not for a set of random variables". Good grief!!! All I want to know is whether or not each functionally dependent value has its own variance. Without smoke and mirrors!
Kind regards, Jan W Merks |