From: Mike Lorrey (mlorrey@yahoo.com)
Date: Thu Oct 31 2002 - 18:51:00 MST
--- "Eliezer S. Yudkowsky" <sentience@pobox.com> wrote:
> mlorrey@datamann.com wrote:
>
> > The probability of what you ask is 50%. The prob that any given
> flip is one or
> > the other is 50% in any situation. The odds that every other flip
> IN SEQUENCE
> > is tails is dependent upon how long your sequence is. In the case
> of getting
> > five tails, the odds are 3.125%.
>
> Thanks, Mike, but this answer is incorrect. See the description of
> the problem. If a *random* sampler reports five heads, the other
> coinflips have a 50% chance of being heads. If a *biased* sampler
> reports five heads, that's an entirely different issue.
Depends on what the bias is. You originally said the bias was in
selecting every other flip. This is not a bias, because the odds for
any given flip is 50%. You are selecting which flips to use before you
know any history, and you are selecting flips in an arbitrary manner.
You might disagree with me. I suggest you run a real experiment before
rejecting it.
__________________________________________________
Do you Yahoo!?
HotJobs - Search new jobs daily now
http://hotjobs.yahoo.com/
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:17:54 MST