commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 29419] - ChiSquareTest does not return
Date Mon, 07 Jun 2004 02:33:44 GMT
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=29419>.
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=29419

ChiSquareTest does not return





------- Additional Comments From phil@steitz.com  2004-06-07 02:33 -------
The ChiSquare test statistic in this case is 3624883 (rounded).  With df = 7,
the p-value returned should be effectively 0.  The code actually does terminate,
when Gamma.regularizedGammaP throws a ConvergenceException.  It takes a *long*
time, however, since the default iteration bound of Integer.MAX_VALUE is used.

It is not obvious to me how to address this.  Reducing the default max
iterations above might be a good idea in general; but this would just cause us
to throw sooneer in the present case, when we should be returning 0.  I am
inclined to suggest that we force 0 for the p-value (or 1 in the ChiSquare cdf)
when the value of the test statistic is greater than say, 1000 * df (which is
500 times the variance) or some such bound.  I will run some tests to see what a
reasonable bound might be.

---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org


Mime
View raw message