Two Plus Two Older Archives

Two Plus Two Older Archives (http://archives2.twoplustwo.com/index.php)
-   Science, Math, and Philosophy (http://archives2.twoplustwo.com/forumdisplay.php?f=45)
-   -   Frequentists and Opponents of Bayes (http://archives2.twoplustwo.com/showthread.php?t=408268)

Gammonster 12-31-2005 03:08 PM

Frequentists and Opponents of Bayes
 
Classes are finally over for me, and I'm picking up my study of inductive reasoning. While I'll be tackling Baye's theorem next, and I'll have questions about that, could anyone recommend a frequentist or two, or any other philosopher or writer opposed to Bayes?

Grisgra 12-31-2005 03:42 PM

Re: Frequentists and Opponents of Bayes
 
I thought that the frequentist/Bayes distinction was limited to statistical methodology . . . I don't believe that philosophy-wise, there is any real challenge (or could be) to Bayes. All judgements should be based partially, when possible, on prior information, after all.

atrifix 12-31-2005 08:08 PM

Re: Frequentists and Opponents of Bayes
 
Bayes's theorem is a theorem, after all, so it has to be true, but it's so simplistic that it's essentially trivial. The interesting thing about Bayes's theorem is that it indicates we should update our expectations based on additional partial knowledge.

The question that concerns philosophers is how should we approach epistemology and philosophy of science? This problem really originates with Hume's problem of induction. The Bayesian response is to simply use probabilities to determine the likelihood of events, and update those probabilities. The obvious objection is that without knowledge of the priors (which Bayesians assume are just given, as if we are omnisicient), we have no idea how to update the posteriors.

Try http://plato.stanford.edu/entries/ep...gy-bayesian/#5 for more. Just check out some of the references in the bibliography for further information.


All times are GMT -4. The time now is 08:02 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, vBulletin Solutions Inc.