Re: Probabilities




On 8-Oct-2008, ilanpsi@xxxxxxxxx wrote:

On Oct 8, 3:40 am, ronaldo_jeremiah <ronaldo_jerem...@xxxxxxxxx>
wrote:
On Oct 7, 7:43 pm, ilan...@xxxxxxxxx wrote:



On Oct 8, 1:09 am, ronaldo_jeremiah <ronaldo_jerem...@xxxxxxxxx>
wrote:

Likelihood that Bigfoot exists =
likelihood that God exists =
likellihood that LANCE won seven TdF without doping.

Thanks,

-rj
Lemon Club VP and
Ordained Minister, Church of the FSM

P.S.  If I'm wrong and those probabilities are in fact unequal, it
could really only be because the first one is nonzero.

It's an interesting question, because it shows that people who don't
know much about probability theory still have an intuitive notion of
what probability should be even though they don't understand. In
partijcular, you question assumes that a probability exists for such
things, not at all clear to my heightened awareness of such things.

-ilan

Ilan -

Your post is a brilliant counter-troll.  I'm actually going to ruin my
original troll to respond, in a serious way, to what you have written
here.

At first I didn't understand what you were saying.  Maybe I still
don't, but I think that I might.  Are you saying that the notion of
'probabilities' doesn't really apply, because those things either are,
or are not, true, and it's just that we don't know them?

Does it substantively change things if I said:

Likelihood we will find Bigfoot =
Likelihood we will confirm god's existence?

-rj

Probability is fairly difficult to define and is somewhat circular.
There are also a number of different interpretations of what it means.
For a repeatable experiment, most people would say that a probablity
of 40% of an event happening, is that in the long run, the event will
happen about 40% of the time. For a one time event, then that won't
work, so the meaning might mean "in all different ways things can turn
out, the event happens in 40% of them". In practice, you assume that a
probability distribution exists and then try to figure out what it is
through statistical analysis of experimentation. On the other hand,
you can never be sure that the probability distribution exists.

-ilan

Why it is much better to rely on an Information Theoretic or Bayesian
approach.

Rick
.



Relevant Pages

  • Re: The danger of classical hypothesis and significance tests [was Re: MADLY AMUSED]
    ... Using the likelihood alone will not guarantee this fact necessarily ... I have seen this attempt before to turn the Bayesian bug into a feature. ... Certainly there is no in-principle necessity to resort to a prior to achieve the same effect. ... The Bayesian bug is the very same that bothered Rev. Thomas Bayes from the beginning, namely the lack of justification for treating the parameter constant as a random variable in its own right, and rewriting the original model as conditional probability. ...
    (sci.stat.math)
  • Re: Coin flipping
    ... a pretty good idea by using a GAM to smooth ... the probability of a head (i.e. you plot H or T against the run number, ... the likelihood is beta distributed, and the point estimate is the proportion of tails. ...
    (sci.stat.math)
  • Re: @ Beldin - probability and likelihood
    ... likelihood of them matching the known probabilities. ... The 'probability of them matching expected values" is a statistical term, ... That has nothing to do with the expected likelihood of the event. ...
    (rec.gambling.poker)
  • Re: urn model
    ... I am having a hard time writing the likelihood of the following urn ... probability of each in terms of p, and see where that takes you. ... I just use the likelihood function to find the MLE of p ...
    (sci.stat.math)
  • Re: Probabilities
    ... Ordained Minister, Church of the FSM ... what probability should be even though they don't understand. ... At first I didn't understand what you were saying. ... Likelihood we will find Bigfoot = ...
    (rec.bicycles.racing)