| |
BGonline.org Forums
Confidence Interval Overlap vs Joint Standard Deviation (long)
Posted By: Bob Koca In Response To: Confidence Interval Overlap vs Joint Standard Deviation (long) (neilkaz)
Date: Saturday, 31 October 2009, at 7:30 p.m.
"What we see here is that if the two 95% CI's don't overlap or if they even touch we have at least 99% confidence. "
No.
Your calculations were assuming that the std dev for each play individually was the same. Suppose though that we have already rolled out play A a LOT and have equity of .60 with s.d. of .001. We then start rolling out play B and after a while we have equity of .65 with std dev. .024 The 95% CI do not overlap but the joint CI is just a little bit more than 95%.
I also point out that the calculation of joint SD makes an independence assumption which I don't think is warranted.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.