Mailing-List: contact commons-dev-help@jakarta.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Jakarta Commons Developers List" <commons-dev@jakarta.apache.org>
Date: 9 Mar 2004 06:20:24 -0000
Message-ID: <20040309062024.25959.qmail@nagoya.betaversion.org>
From: bugzilla@apache.org
To: commons-dev@jakarta.apache.org
Cc: 
Subject: DO NOT REPLY [Bug 27524]  -
    Percentile calculation is not correct

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=27524>.
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=27524

Percentile calculation is not correct


------- Additional Comments From phil@steitz.com  2004-03-09 06:20 -------
Excel is using a different percentile interpolation definition than what [math]
currently uses. Unfortunately, the javadoc and test cases describing the [math]
algorithm have been inadvertently removed. [math] uses the "index number" method
with simple linear interpolation as described in the first example here:

http://www.math.bcit.ca/faculty/david_sabo/apples/math2441/section4/relstanding/RelStanding.htm

The code is working as designed.

The reference above also describes the method used by Excel, which is another
commonly used algorithm that gives different results.  R uses this method as well.

I am +1 to changing to the R / Excel method and will make the change (and add
complete javadoc) if there are no objections.

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