home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!darwin.sura.net!paladin.american.edu!auvm!CCWF.CC.UTEXAS.EDU!RCOLE
- Message-ID: <9301062032.AA20180@sylvester.cc.utexas.edu>
- Newsgroups: bit.listserv.stat-l
- Date: Wed, 6 Jan 1993 14:32:16 -0600
- Sender: STATISTICAL CONSULTING <STAT-L@MCGILL1.BITNET>
- From: "Richard W. Cole" <rcole@CCWF.CC.UTEXAS.EDU>
- Subject: contingency tables
- Lines: 13
-
- another way to generate comparisons among rows or comparisons
- among columns for contingency tables is to use Correspondence Analysis
- where each row is defined as a point in a weighted Euclidean metric
- of dimensions (min(r,c)-1). This is a chi-squared metric where
- distances among points can be considered a single degree of freedom
- chi-square test. More interesting is the notion of confidence
- ellipses around each point and the origin of the space.
-
- ========================================================
- Richard Cole, Stat/Math Services Learning to
- Computation Center UT Austin TX, 78712 Paint by
- rcole@ccwf.cc.utexas.edu Numbers
- ========================================================
-