| |
BGonline.org Forums
Confidence Interval Overlap vs Joint Standard Deviation (long)
Posted By: Tom Keith In Response To: Confidence Interval Overlap vs Joint Standard Deviation (long) (neilkaz)
Date: Saturday, 31 October 2009, at 5:51 p.m.
To me, the important question is: How likely is it that the results will change if I continue this rollout with more trials?
I like the JSD feature of GnuBG because it gets you closer to answering that essential question. Even better would be a bot that tells you directly: "If you continue this rollout indefinitely, there is a 1.7% chance that a different candidate will come out on top."
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.