Re: Fuzzy logic compared to probability

Josh Kuperman (josh@racing.saratoga.ny.us)
Fri, 1 Mar 1996 21:07:04 +0100


I am a CS student who is trying to use horse-race betting as a way to
understand a variety of problems with decision making and AI. I have found
that it seems to require a better background in statistics, especially
Bayesian statistics than I have. That aside, while I'm not really
contributing anything to this thread I have a few questions about some
peripheral comments.

In article <4gnr70$1l6v@b.stat.purdue.edu>, hrubin@b.stat.purdue.edu
(Herman Rubin) wrote:

> In article <4gn6re$r5u@watt.electriciti.com>,
> Fred A Watkins <fwatkins@hyperlogic.com> wrote:
> >hrubin@b.stat.purdue.edu (Herman Rubin) wrote:
> >>In article <312B60FB.41C67EA6@colorado.edu>,
> >>Robert Dodier <dodier@colorado.edu> wrote:
[etc]
>
> >>probability from utility. See my paper in _Statistics and
> >>Decisions_, 1987.
Did you post a source for that paper. If so could you email me where to
find it. The beginnings of the thread aren't on my server anymore.
>
>
> >The trouble with probability as "degree of belief" is that if there
> >were no people, there would be no belief, hence no probability. And
> >it's likely that one's belief is ultimately grounded in experience,
> >anyway.

That argument strikes me as very odd as the "if a tree falls in the forest
.. " argument strikes me as odd. Afterall, without an observer there'd be
no observations of frequency either.
>
>
> >>>In particular, can one bet on a `degree of truth' ??

As a gambler let me note you can bet on anything as long as you have
someone to hold the bet, someone to lay the odds, someone to take the
odds, and a way to settle the bet. As for degree of truth - I don't think
a bookie would touch it.

> This is the "Dutch Book" theorem. It only uses algebra. It does
> not require knowing anything other than algebra.

Can anyone tell me where "Dutch Book" derives. I know that in the
bookmakers ring a long time ago a Dutch Book was when someone offered
fair-odds and the other bookmakers would put him out of business by
betting against his book. I also know that in horse-racing circles
"dutching systems" are those where you bet on more than one (or all) of
the possible outcomes.
>

-- 
Josh Kuperman
josh@racing.saratoga.ny.us
http://racing.saratoga.ny.us/~josh (barely under contruction)