commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Al Chou <hotfusion...@yahoo.com>
Subject RE: [math] Only sample variances?
Date Mon, 16 Aug 2004 03:00:33 GMT
--- Brent Worden <brent@worden.org> wrote:
> Any objection to tracking these and future feature requests using the wiki?
> A wish list, if you will.

+1

We should formally note somewhere that that's the procedure we follow.


Al


> -----Original Message-----
> From: Phil Steitz [mailto:phil@steitz.com]
> Sent: Sunday, August 15, 2004 1:06 PM
> To: Jakarta Commons Developers List
> Subject: Re: [math] Only sample variances?
> 
> 
> Kim van der Linde wrote:
> > Hi,
> >
> > I just looked through the variance class, and saw that
> > it only supports sample vasriance (N-1) but not
> > population variances (N). Any reason for this, and can
> > it be added?
> 
> It would probably be better to add this to one of the aggregates or
> StatUtils or to add a new statistic altother. This is because a
> UnivariateStatistic can only report one value. The most commonly used (at
> least in statistical applications) value for Variance and Standard
> Deviation is the sample statistic, which is why that is what the
> statistics named "StandardDeviation" and "Variance" report. From a user's
> perspective, its not too hard to convert; but I agree that it would be
> convenient to add these. Probably best to make new statistics in the
> moment subpackage called "PopulationXxx".  Patches welcome ;-)
> 
> Phil
> 
> 
> >
> > Cheers,
> >
> > Kim

---------------------------------------------------------------------
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