home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #27 / NN_1992_27.iso / spool / bit / listserv / statl / 2000 < prev    next >
Encoding:
Text File  |  1992-11-17  |  1.3 KB  |  30 lines
  1. Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
  2. Path: sparky!uunet!stanford.edu!bcm!convex!darwin.sura.net!wupost!gumby!yale!yale.edu!news.yale.edu!mars.caps.maine.edu!maine.maine.edu!cunyvm!psuvm!auvm!AXE.HUMBOLDT.EDU!CHAMBERLINC
  3. X-VMS-To: IN%"stat-l@vm1.mcgill.ca"
  4. Message-ID: <01GR956K54OG8WWIT8@AXE.HUMBOLDT.EDU>
  5. Newsgroups: bit.listserv.stat-l
  6. Date:         Tue, 17 Nov 1992 09:40:00 PST
  7. Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
  8. From:         Charles Chamberlin <CHAMBERLINC@AXE.HUMBOLDT.EDU>
  9. Subject:      Re: Normal probability plots
  10. Lines: 18
  11.  
  12. Shapiro,Wilk,and Chen(1968) JASA 63:1343-1372 report an empirical
  13. comparison of alternative methods for testing for normality.  The
  14. methods they compare include the chi-squared test, the Kolmogorov-
  15. Smirnov test, as well as the Shapiro-Wilk test. 45 alternative
  16. distributions in 12 families and five sample sizes were studied.
  17. Overall, the S-W test was most powerful.
  18. It also can be applied to small sample sizes.
  19.  
  20. -------------------------------------------------------------------------------
  21. Charles E. Chamberlin
  22. Environmental Resources Engineering
  23. Humboldt State University
  24. Arcata, CA  95521
  25. USA
  26. (707) 826-4346
  27. (707) 826-3616 (fax)
  28. ChamberlinC@axe.humboldt.edu
  29. -------------------------------------------------------------------------------
  30.