-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 07/31/07 11:08, Johannes Wiedersich wrote: > [snip] >>> Because people (even, apparently, smart people like you) seem to fixate on >>> that "average" number and think that "because that's the average, that's >>> what the temperature is supposed to be", weather sites should *not* post >>> average, but *should* post the *range* of temperatures 1 STDDEV above and >>> below the average. >> please Ron. I'm fully aware of what an average is and was merely jumping on >> the "let's >> discuss random weather stuff" band-wagon and having a little fun. > >> sheesh. > >> I agree though that the range of temperatures would be really nice to >> see. The high and low with 1 stddev above and below each would >> actually be useful information. > > Sorry for joining this OT thread. > > Don't get misled by the subject; this has nothing to do with mathematics: > > 1) stddev's only apply, if temperatures would vary statistically and
By "statistically", do you mean "evenly"? (It's been a number of decades since I've needed to know more than just the generalities of bell curves, standard deviations, tails, etc.) > would therefore show a Gaussian distribution around a mean value. > Neither is the case. > > 2) If temperatures would vary statistically, it would be enough to give > the stddev, because 'above' and 'below' would be the same value. > > The fact that 'above' and 'below' are generally different values, is a > direct hint that 1) does not apply. This is probably one of the reasons > why meteorologists give mean temperatures in January and July, but no > stddev. I don't know what the year-on-year temperature variability is like in Germany, but here in the US it's quite high. - -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFGsGNBS9HxQb37XmcRAsUEAKCreHvA+ivRSeSIr5ElxBuipeYrKwCgmEpJ qi0BYqWapdcP5eVB40DUh/k= =HekY -----END PGP SIGNATURE----- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
