
BGonline.org Forums
Statistical C.I. Question
Posted By: Daniel Murphy
Date: Thursday, 29 January 2015, at 3:21 p.m.
In Response To: Statistical C.I. Question (sita)
A useful site!
For what it's worth, if you're testing for frequency of doublets (which should be 16.67%), the 95% confidence interval for only 36 rolls is ±12.17, for a range of 4.5% to 28.84%. To cut the CI in half, you need to quadruple the number of trials. Thus,
Rolls 95% CI 36 ±12.17% 144 ±6.09% 576 ±3.04% ... 4,608 ±1.08% ... 589,824 ±0.1% Keep rolling, often and well.

BGonline.org Forums is maintained by Stick with WebBBS 5.12.