home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!IRLEARN.UCD.IE!RCONROY
- Message-ID: <STAT-L%92111604173290@VM1.MCGILL.CA>
- Newsgroups: bit.listserv.stat-l
- Date: Mon, 16 Nov 1992 09:12:49 GMT
- Sender: "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
- From: Ronan M Conroy <RCONROY@IRLEARN.UCD.IE>
- Subject: Normal probability plots
- Lines: 13
-
- I frequently use normal probability plots to check distributions
- of variables visually. I also have the facility (on Data Desk)
- to get a correlation between the data and the normal distribution.
- Does anyone have information on whether there's much you can
- tell from the correlation? My own experience is that you can get
- an r of > 0.9 from the log of a normal distribution, and the
- ones that look good by eye usually have an r in the range 0.98
- upwards. Twisting the tails of the plot, however, where data are
- thin, isn't going to affect the r a whole lot, but is going to
- affect your distribution. Maybe r is simply anecdotal...
- Any views?
-
- ronan
-