My only quarrel with you analysis is that the null hypothesis is that the two distributions are equal, not that they are unequal. So because you want to accept the null hypothesis, it is unwise to interpret the result in terms of confidence. That interpretation applies when you wish to reject the null. It's been a long time since I thought about this so please take that into consideration.
-- (B=) ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
