home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!CUNIXB.CC.COLUMBIA.EDU!JSS19
- Message-ID: <CMM.0.90.0.714430897.jss19@cunixb.cc.columbia.edu>
- Newsgroups: bit.listserv.sas-l
- Date: Fri, 21 Aug 1992 17:01:37 EDT
- Reply-To: Jill S Shapiro <jss19@CUNIXB.CC.COLUMBIA.EDU>
- Sender: "SAS(r) Discussion" <SAS-L@UGA.BITNET>
- From: Jill S Shapiro <jss19@CUNIXB.CC.COLUMBIA.EDU>
- Subject: Mahalanobis distance
- Lines: 19
-
- I was hoping that someone might be able to help with a question
- concerning discriminant analysis and Mahalanobis generalized
- distances. The problem is as follows: In SAS I can obtain the
- distances but not their significance levels while in SPSS I can
- obtain the F-Statistics and significance levels for the distances
- which are calculated internally but which SPSS can't print out.
- I'm trying to find someway of getting one package to provide both
- pieces of data as I don't feel comfortable combining the separate
- results. For those who usually work in SAS and don't get
- significance levels is there a general rule of thumb that you
- follow in evaluating these distances?
- Any suggestions/comments would be appreciated.
- Thank you.
-
-
- I can be contacted at
- JSS19@CUNIXB.CC.COLUMBIA.EDU
- or
- JISUT@CUVMB.BITNET
-