home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #23 / NN_1992_23.iso / spool / bit / listserv / statl / 1708 < prev    next >
Encoding:
Text File  |  1992-10-10  |  847 b   |  20 lines
  1. Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
  2. Path: sparky!uunet!paladin.american.edu!auvm!CCDDCD1.EM.CDC.GOV!MMZ1
  3. Encoding: 8 TEXT
  4. X-Mailer: Microsoft Mail V3.0 (beta-2)
  5. Message-ID: <2AD6782D@router.em.cdc.gov>
  6. Newsgroups: bit.listserv.stat-l
  7. Date:         Sat, 10 Oct 1992 10:27:00 EST
  8. Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
  9. From:         mmz1@CCDDCD1.EM.CDC.GOV
  10. Subject:      Disciminant analysis with dichotomies?
  11. Lines: 7
  12.  
  13. Logistic regression can use both dichotomous and continuous variables for
  14. discrimination and should be applicable to your situation.  In comparisons
  15. with discriminant analysis, logistic regression does worse only when all the
  16. independent variables are normally distributed; otherwise, it does as good or
  17. better than linear discriminant analysis.
  18.  
  19.                                     Matthew Zack
  20.