[19026] in s-news-athena
[S] Comparison of Chisquare p-values which close to zero
daemon@ATHENA.MIT.EDU (John Lee)
Tue Aug 24 12:28:21 1999
Message-Id: <19990824161403.3991.qmail@hotmail.com>
From: "John Lee" <statcomp@hotmail.com>
To: s-news@wubios.wustl.edu
Date: Tue, 24 Aug 1999 11:14:03 CDT
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Dear Splusers,
I have a general question about the p-value comparison.
We want to select the best model according to Chi-square
test among the several 2-way contingency table. Problem
is that we got many tables whose p-value literally equals to zero.
So we are now thinking about chi-square value divided by degrees
of freedom for selecting best model. But it can be a problem
with small degrees of freedom. Do you have any other idea
or suggestion?
Thanks
John Lee
_______________________________________________________________
Get Free Email and Do More On The Web. Visit http://www.msn.com
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
