Re: AI-GEOSTATS: Calculating averages
Hi Craig The average of a product will only equal the product of the averages if the two numbers are completely independent of one another. This is exactly analogous to the calculation of a covariance in statistics. Isobel http://uk.geocities.com/drisobelclark __ Do You Yahoo!? Everything you'll ever need on one web page from News and Sport to Email and Music Charts http://uk.my.yahoo.com -- * 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
Re: AI-GEOSTATS: Original Covariogram paper.
Digby There is a 96 page bibliography in the back of Cressie's book! Cressie, Noel A. C., Statistics for Spatial Data, Revised Edition, 1993, John Wiley & Sons, Inc, New York, 900pp. The earliest paper I know of about variograms (but not using that terminology) is by Prof de Wijs in a paper in the SAIMM journal around 1953. He called it studying "successive differences" and wrote it whilst on sabbatical in South Africa. Hence the "de Wijsian model" cited by Matheron - who did coin the terminology in 1963. There is also a paper by Cressie in Math Geol called "The Origins of Kriging" which came out in the early 90s. Isobel http://geoecosse.bizland.com/news.html __ Do You Yahoo!? Everything you'll ever need on one web page from News and Sport to Email and Music Charts http://uk.my.yahoo.com -- * 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
Re: AI-GEOSTATS: Calculating averages
Calculating averagesCraig, Don't you mean multiply Average Nickel Head grade by Average Mill recovery. Your first figure 0.3342 is correct but your figure 0.315963265 is incorrect. It is your assumption that your average mill recovery is tonneage recovery, i.e. you averaged the mill recovery as if it where recovery per tonne of ore instead of per % of metal so your calculation of 0.5557*0.56857 is incorrect. Your average mill recovery should be calculated (0.53*60)+(0.65*62)+(0.32*20)+(0.8*70)+(0.56*63)+(0.45*55)+(0.58*0.68)/ (0.53+0.65+0.32+0.8+0.56+0.45+0.58) =0.601465 So your original table should look like; Nickel MillRecovered HeadRecoveryNickel Grade % (Ni %) 0.5360 0.3180 0.6562 0.4030 0.3220 0.0640 0.8 70 0.5600 0.5663 0.3528 0.4555 0.2475 0.5868 0.3944 Average 0.555714286 60.14652956 0.3342 Average Nickel Head Grade divided by Average Mill Recovery > 0.334242857 Regards Digby Millikan Geolite Mining Systems [EMAIL PROTECTED] http://www.users.on.net/digbym -- * 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
AI-GEOSTATS: Calculating averages
Title: Calculating averages Hello, I was hoping advice could be given on the following problem. We store nickel (eg 0.40 Ni %) and mill recovery (55.45%) in our mine model. From this we calculate a mill recovered nickel (eg 0.22 RecNi %) and store this in the model. All 3 model items are stored to two decimal places. The rounding is to three decimal places (i.e. +/-0.005). We create ore-cuts based on the tonnes-weighted average of the Mill Recovered Nickel (using an Ore-material classification scheme based off Mill Recovered Nickel). My problem is given below and highlighted in green. I have assumed tonnage weighting in each row is 1. Nickel Mill Recovered Head Recovery Nickel Grade % (Ni %) 0.53 60 0.3180 0.65 62 0.4030 0.32 20 0.0640 0.8 70 0.5600 0.56 63 0.3528 0.45 55 0.2475 0.58 68 0.3944 Average 0.555714286 56.85714286 0.3342 Average Nickel Head Grade divided by Average Mill Recovery > 0.315963265 It would appear that the averaging of the recovered nickel (0.3342) gives a different result to dividing the average Nickel Head Grade by the Average Mill Recovery (0.315963265). These averages are generated in Microsoft Excel so it doesn't seem to be a rounding error. We believe the first average is wrong and it would appear that the greater the variability of the data set, the greater the difference between the two calculation averages. If possible, would anyone know what is happening here and which is the correct approach. Thank-you in appreciation. Craig Allison MKO Resource Evaluation Geologist