DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG· RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://issues.apache.org/bugzilla/show_bug.cgi?id=36215>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND· INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=36215 ------- Additional Comments From [EMAIL PROTECTED] 2005-08-17 20:36 ------- I think we're witnessing the accumulation of numerical error. The cumulative probability method uses the textbook definition for its computation: summing the point probabilities. Each of these probabilities results in a small numerical error and the process of summing them just exacerbates the error. In your case, X ~ hyper(2000, 500, 100) and P(X <= 50) there are 51 small errors that, when added together, cause the total sum to be over 1.0. Also, the point probabilities the larger x values (30 - 50) are so small that they provide little value to the overall sum, just error. I think we can address this problem by simply short-circuiting the summation whenever it goes above 1.0. This will solve the problem of cumulative probablities greater than 1 and make the upper tail probabilties correct because, again in your case, P(X >= 50) is extremely close to zero. What do others think? -- Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
