Who knows? Munk has always been driven to mine gold! So his offspring may be driven too! Do Barrick’s investors worry? Who cares? Here are some facts and figures! Once upon a time I had asked Barrick Gold Corporation to invest C$100,000 in applied statistics. What I had been given was a massive set of gold grades in drill core samples. I converted gold grades into masses of in-situ gold. Next, I derived 95% confidence limits for gold grades and contents. I have not yet described in detail how I unscrambled the Bre-X salting fraud! Dr Roussos Dimitrakopoulos is sticking to stringing Markov chains rather than applying a simple test for spatial dependence between measured values in ordered sets and counting degrees of freedom. Who could possibly be behind stringing Markov chains? Stay tuned to find out why truth is stranger than fiction.
Barrick Gold has awarded C$1million to Dr Roussos Dimitrakopoulos. Dr RD prefers to ignore degrees of freedom and to dismiss real variances. He prefers to string Markov chains between measured values in ordered sets. He cannot possibly derive unbiased confidence limits for gold grades and contents. But Dr RD keeps happily stringing along! McGill’s students have not yet been taught how to count degrees of freedom and why! Of course, it makes no sense whatsoever to string functionally dependent values between measured values in ordered sets. What makes sense is testing for spatial dependence between gold grades within boreholes! Stuffing strings of Markov chains between measured values makes no sense whatsoever!
My memo on Statistical Evaluation of Test Results was sent to Mr Alan R Hill, Executive President, on December 17, 1996. I was pleased to have so many test results for gold. What lingers in my mind is a visit to Barrick’s office in Toronto. I had been invited to come by on December 17, 1996 at 10:00AM. We were talking about tasks I thought ought to be done. That’s when the meeting room door opened ajar. It was Chairman Munk! All he did was ask President Carrington to come and see him after we were finished.
I had been talking about a test program at Bre-X’s site in the Kalimantan jungle. I was given 2,200 test results in crushed core samples each with a length of 2.9 m. Lakefield Research had measured gold by atomic absorption in what was called 0.1 m library core. I wondered what could have been added to 2.9 m of crushed core samples! Filings of a gold/silver alloy?
Several books have been written about Bre-X and its turbulent existence. I have a copy of each. Gold Today, Gone Tomorrow squeezed more facts between its covers than others did. Who keeps track of those who strip variances off distance-weighted averages AKA kriged estimates? Wikipedia published goofy science under Geostatistics. All it took was to strip the variance off the distance-weighted average and call what’s left a kriged estimate. Geostatistocrats like to play kriging games with infinite sets of kriged estimates and zero kriging variances. Good grief!