ellis@spar.UUCP (Michael Ellis) (04/29/85)
Comments on several of Geffrey Clemm's recent remarks -- >> When science labels this behavior as random, all that means is that >> it is beyond the ability of science to predict what it is going to do.. > >This is by no means all that it means. One of the central purposes of >the science of statistics is to differentiate between pointlessly random >(spiritually random ??) behaviour and between some kind of connection between >behaviour, where these terms are given very explicit meanings. Do you mean to imply that statistics will unfailingly discover any and all possible connections? I believe it would be child's play for a suitably trained mathematician to produce data that would be labeled `random' by any statistical analysis in existence, yet said data would contain deeply organized underlying structure. >> Unrecognized organization is frequently perceived as `random', yes? >No. My! Such a fountain of wisdom!! Forgive my impertinence, but I disagree with this argument, ingenious though it may be... Oh yes, where was I? The digits of pi, for example, are random by any statistical analysis I know. But it only appears random when it is viewed as a sequence of decimal digits -- if you see it as 4 * (1 - 1/3 + 1/5 - 1/7 + 1/9 ...), its randomness disappears. Unrecognized organization is frequently perceived as `random', yes? -michael