Difference between revisions of "Monte Carlo simulations"
(Created page with "Monte Carlo simulations are used to explore variability in a data set that is Normally distributed (a.k.a. Gaussian). This distribution is a fundamental part of statistics an...") |
|||
| Line 1: | Line 1: | ||
| + | [[Category:Monte Carlo]] |
||
Monte Carlo simulations are used to explore variability in a data set that is Normally distributed (a.k.a. Gaussian). This distribution is a fundamental part of statistics and shouldn't be confused with "commonly distributed". Some grindability test results are expected to be Normally distributed (such as work index, Mia) and other are not Normally distributed (Axb) and are not suitable for simulation. |
Monte Carlo simulations are used to explore variability in a data set that is Normally distributed (a.k.a. Gaussian). This distribution is a fundamental part of statistics and shouldn't be confused with "commonly distributed". Some grindability test results are expected to be Normally distributed (such as work index, Mia) and other are not Normally distributed (Axb) and are not suitable for simulation. |
||
Revision as of 19:15, 25 May 2023
Monte Carlo simulations are used to explore variability in a data set that is Normally distributed (a.k.a. Gaussian). This distribution is a fundamental part of statistics and shouldn't be confused with "commonly distributed". Some grindability test results are expected to be Normally distributed (such as work index, Mia) and other are not Normally distributed (Axb) and are not suitable for simulation.
The Monte Carlo engine in SAGMILLING.COM acts using synthetic test samples that define the proportions and distribution parameters for performing the simulations. **It is highly recommended to run Monte Carlo simulations in their own project** and do not mix synthetic Monte Carlo test data with actual laboratory test results in the same project.
It is also strongly recommended to mark all synthetic Monte Carlo test samples with the Synthetic flag set to 1 so that the jibberish data of the Monte Carlo samples does not affect the Testwork Comparison plots where you compare two test results in a 2-D plot.
