home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!uvaarpa!darwin.sura.net!bogus.sura.net!howland.reston.ans.net!paladin.american.edu!auvm!AXE.HUMBOLDT.EDU!CHAMBERLINC
- X-VMS-To: IN%"STAT-L@VM1.MCGILL.CA"
- Message-ID: <01GU09HNIU8G8WW63X@AXE.HUMBOLDT.EDU>
- Newsgroups: bit.listserv.stat-l
- Date: Wed, 27 Jan 1993 08:32:00 PST
- Sender: STATISTICAL CONSULTING <STAT-L@MCGILL1.BITNET>
- From: Charles Chamberlin <CHAMBERLINC@AXE.HUMBOLDT.EDU>
- Subject: RE:testing for periodicity
- Lines: 15
-
- Data that may contain periodic components are often best characterized
- by the use of circular distributions. These are suitable for interval
- scales of measurement where there is no non-arbitrary zero point and no
- non-arbitrary high/low values. Observations with respect to direction,
- time-of-day, day-of-year, and day-of-lunar_cycle ( and in some cultural
- settings day-of-week) are commonly described using circular distributions.
- Zar[, J.H. (1984) Biostatistical Analysis, 2nd ed., Prentice-Hall]
- provides a good introduction to the use of curcular distributions in
- chapters 24 and 25.
- ---------------------------------------------------------------------
- Charles E. Chamberlin (707) 826-4346
- Environmental Resources Enginering (707) 826-3616 (fax)
- Humboldt State University InterNET:
- Arcata, California 95521 chamberlinc@axe.humboldt.edu
- ----------------------------------------------------------------------
-