Bob O'H
Posts: 2564
Joined: Oct. 2005
|
I'm sorry to say, but Allen is embarrassing himself too. Unless one is very strict, this is wrong:
| Quote | | First of all, a Gaussian distribution is not “the result of a random process in which small steps are taken in any direction”. |
(there are other ways of producing a Gaussian distribution, including the Monty Python method). But if you have a random walk with small steps, the distribution becomes Gaussian. It's a consequence of the central limit theorem(s). Another consequence is Fisher's work where he combined Darwin and Mendel.
Sheldon's mistake is to think that the only way to be random is to be normal. He then equates randomness in traits to randomness in time to events, without explaining why this isn't utter rubbish.
Luckily, Mark Frank knows what he's talking about: as he points out, if events are random (with a constant rate), you get an exponential distribution for the time to the events.
--------------
It is fun to dip into the various threads to watch cluelessness at work in the hands of the confident exponent. - Soapy Sam (so say we all)
|
