Difference between revisions of "Thread:Talk:RoboRunner/calculating confidence of an APS score/reply"
Jump to navigation
Jump to search
Skilgannon (talk | contribs) (Reply to calculating confidence of an APS score) |
Skilgannon (talk | contribs) (fix link) |
||
Line 1: | Line 1: | ||
− | I don't actually think this can be correctly modelled by a unimodal distribution - you will be adding thin gaussians to fat gaussians, making horrible bumps which don't like to be approximated by a single gaussian mean+stdev. I almost wonder if some sort of [ http://en.wikipedia.org/wiki/Monte_Carlo_method Monte-Carlo] solution wouldn't be most accurate in this instance - at least the math would be easy to understand. | + | I don't actually think this can be correctly modelled by a unimodal distribution - you will be adding thin gaussians to fat gaussians, making horrible bumps which don't like to be approximated by a single gaussian mean+stdev. I almost wonder if some sort of [http://en.wikipedia.org/wiki/Monte_Carlo_method Monte-Carlo] solution wouldn't be most accurate in this instance - at least the math would be easy to understand. |
Latest revision as of 23:12, 13 August 2012
I don't actually think this can be correctly modelled by a unimodal distribution - you will be adding thin gaussians to fat gaussians, making horrible bumps which don't like to be approximated by a single gaussian mean+stdev. I almost wonder if some sort of Monte-Carlo solution wouldn't be most accurate in this instance - at least the math would be easy to understand.