home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Subject: (no subject given)
- Path: sparky!uunet!paladin.american.edu!auvm!IDUI1.BITNET!KIRK
- Message-ID: <STAT-L%92111916531719@VM1.MCGILL.CA>
- Newsgroups: bit.listserv.stat-l
- Date: Thu, 19 Nov 1992 16:53:17 -0500
- Sender: STATISTICAL CONSULTING <STAT-L@MCGILL1.BITNET>
- From: KIRK@IDUI1.BITNET
- Lines: 16
-
- There has been some interest in information on chi-square goodness-
- of-fit vs. edf vs. Shapiro-Wilks tests for normality. For those of
- you who do not want to spring for the price of the d'Agostino book,
- look at the article by
-
- M. A. Stevens, EDF statistics for goodness of fit and some comparisons,
- JASA 1974, vol. 69, pp. 730-737.
-
- It is a very nice (and readable) paper. As alluded to by others, it
- shows that chi-square tests are passe and that the Shapiro-Wilks test
- is the test of choice for testing normality. For other distributions
- where Shapiro-Wilks does not apply, various EDF tests beat the chi-
- square test hands down. I heartily recommend this paper.
-
- Kirk Steinhorst, University of Idaho
- kirk@laplace.mathstat.uidaho.edu
-