>let's say we have SAT scores where the mean is about 500 and the SD about >100 ... here ... the COV is 100/500 = 20% > >compared to another case where the data are temperatures ... where the SD = >5 (and it's not even this in honolulu!) and the mean is 80 ... so the COV = >5/80 = 6% > >for the life of me, i have never seen how this comparison of 20% versus 6% >has ANY useful meaning > >does anyone out there?
comparing the means from those two examples doesn't have much useful meaning either, that doesn't imply the means aren't useful. I have used CV in comparision of poker situations, where aggresive play will increase your expected return and also your variance. Return distributions in gambling tend to not by symmetric and a small CV helps identify situaitons where you can increase your expected return without a corresponding increase in risk. Probably a better measure in a gambling situation though is Var/mean -- but the conclusions you can draw are the same pretty much whichever you use. Gary Carson . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
