Actual left/right wing discussion Message Board

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Actual left/right wing discussion

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: TimF who wrote (6076)	3/10/2007 4:32:21 PM
From: Lazarus_Long	of 10087

Absolutely, Tim. One of the most difficult areas is experimental design- -to set up an experiment that REALLY tests what you claim it does and doesn't give incorrect results because of some unaccounted-for influence. AND it is quite common in the sciences to simply assume a variable follows a normal (or gaussian or bell-shaped-curve- -all the same thing) distribution, then use the mathematical convenience of that distribution to draw other conclusions. Yet it is known that a great many variables DO NOT follow that curve- -and any conclusions drawn assuming it does are questionable at best. For example, queueing times generally follow a Poisson distribution- -but not always. It is known, for example, that the magnitude of stock price changes is not gaussian- -it has "fat" tails and the frequency of large changes (such as the 500-point-drop we saw recently) is MUCH more likely than a gaussian distribution allows- -orders of magnitude larger. Tests of significance on the matters you mention would probably suggest strongly that the results were due to chance. And there is always the old saw that correlation is not causation.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next