RE: Descriptive statistics

From: Adam Treister (adam@treestar.com)
Date: Thu May 30 2002 - 13:11:14 EST


>  Jan Bayer wrote:
> I think you should consider reading the probability binning papers,
> published sep 2001 (vol 45) in cytometry.
>
>
The important extension is that the probability binning methods include
multivariate
comparisons.

Your points are valid improvements over the well known limitations of the KS
algorithm, but they still look at only one dimension of the data.  This was
exactly how we started when we set out to add comparison statistics to
FlowJo.  But when we realized we should look at all of the data, and
implemented the multivariate comparisons, things worked much better.
Suddenly the software started quantifying the differences in ways that more
closely matched human assessments, and it also became able to gate on the
events that constitute the important differences.

http://www.flowjo.com/v4/html/comparison.html

Adam
------------------------------------------------------------------
Adam Treister
Tree Star, Inc.
ph: 800-366-6045 intl: 1-650-591-2854 fax: 1-650-508-9186
adam@treestar.com    www.flowjo.com
------------------------------------------------------------------



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:26:11 EST