Re: More Hard Problems Using Bayes' Theorem, Please

From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Mon Jul 07 2003 - 21:36:44 MDT

  • Next message: Terry W. Colvin: "FWD [forteana] Re: James Ossuary and Jehoash Inscription are fakes"

    Lee Corbin wrote:
    > Mitchell writes
    >
    >>But what I still don't get is what's wrong with
    >>'orthodox statistics'. I obtained my p1/(p1+p2)
    >>through frequentist thinking, which is supposed to
    >>be anathema to Bayesians. On the SL4 list, Ben
    >>Goertzel implies that "non-Bayesian probabilists"
    >>are restricted in their choice of priors:
    >>http://sl4.org/archive/0305/6820.html
    >
    > I can't answer your question, but did you check out problem
    > number three on my list? Apparently Bayesian statisticians
    > can solve it whereas those like me trained in classical
    > mathematics, do not have enough information. I'm still in
    > shock that not in all instances does P(A & B) = P(A|B)*P(B)
    > according to the Bayesians.

    Eh? Nani? Since when?

    Remember, frequentists are just confused, roundabout Bayesians who
    occasionally give wrong answers - it's not that Bayesians "abjure
    frequentist methods", since, of course, all frequentist methods are merely
    distorted and remixed Bayesian methods. For example, I don't believe the
    common accusation that frequentists have no priors, which is really quite
    an awful thing to say about anyone. Frequentists have priors, they just
    refuse to admit it. Even the most hardened frequentists can become
    Bayesians at any time just by admitting they have priors. Similarly, the
    Bayesian Way derives entirely from the shining light of the Theorem, which
    determines everything quite simply; the Bayesian Way precedes frequentism
    and has nothing to do with what frequentists may or may not do.

    -- 
    Eliezer S. Yudkowsky                          http://singinst.org/
    Research Fellow, Singularity Institute for Artificial Intelligence
    


    This archive was generated by hypermail 2.1.5 : Mon Jul 07 2003 - 21:48:27 MDT