Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.csd.uwm.edu!bionet!ames!uhccux!munnari.oz.au!cs.mu.oz.au!ok From: ok@cs.mu.oz.au (Richard O'Keefe) Newsgroups: comp.lang.pascal Subject: Re: Algorithm for average time of day? Message-ID: <2068@munnari.oz.au> Date: 9 Sep 89 10:47:20 GMT References: <1780003@hpcc01.HP.COM> <898@dutrun.UUCP> Sender: news@cs.mu.oz.au Lines: 14 In article <1780003@hpcc01.HP.COM> azarian@hpcc01.HP.COM (Randy Azarian) writes: >Does anyone have an algorithm available that will calculate >an average time of day? In article <898@dutrun.UUCP>, winfave@dutrun.UUCP (Alexander Verbraeck) writes: > I do not have a readily available algorithm, but I don't think it would > be very difficult writing one. I didn't reply to this earlier because I was sure that someone else would point it out, but "average time of day" is not a well-defined concept. The point is that things like the mean and median are defined for LINEAR scales, but time of day is CYCLIC. Run, do not walk, to the statistics department of your local university, and ask for advice about "location measures for circular distributions". Or post a question in sci.math.stat.