Axel comments on What is the best paper explaining the superiority of Bayesianism over frequentism? - Less Wrong

-1 Post author: Meni_Rosenfeld 01 January 2013 08:58PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (32)

You are viewing a single comment's thread. Show more comments above.

Comment author: JonathanLivengood 02 January 2013 03:34:14AM 12 points [-]

It might help if you told us which of the thousands of varieties of Bayesianism you have in mind with your question. (I would link to I.J. Good's letter on the 46656 Varieties of Bayesians, but the best I could come up with was the citation in Google Scholar, which does not make the actual text available.)

In terms of pure (or mostly pure) criticisms of frequentist interpretations of probability, you might look at two papers by Alan Hajek: fifteen arguments against finite frequentism and fifteen arguments against hypothetical frequentism.

In terms of Bayesian statistics, you might take a look at a couple of papers by Dennis Lindley: an older paper on The Present Position in Bayesian Statistics and a newer one on The Philosophy of Statistics.

Lindley gives a personalist Bayesian account. If you want "objective Bayes," you might take a look at this paper by James Berger. (The link actually has a bunch of papers, some of them discussing Berger's paper, which is the first in the set.)

You might also find Bradley Efron's paper Why Isn't Everyone a Bayesian? useful. And on that note, I'll just say that the presupposition of your question (that Bayesianism is straightforwardly superior to frequentism in all or most all cases) is more fraught than you might think.

Comment author: Axel 02 January 2013 02:09:55PM 2 points [-]

Would this be I.J. Good's letter on the 46656 Varieties of Bayesians? (I'm practicing my google-fu)

Comment author: Pablo_Stafforini 02 January 2013 02:47:41PM *  4 points [-]

That pdf is a scan of chapters 3 and 4 of I. J. Good's book, Good Thinking: The Foundations of Probability and Its Applications (free pdf) (Minnesota: University of Minnesota Press, 1983). Chapter 3, '46656 varieties of Bayesians', reprints a letter in American Statistician (December, 1971), vol. 25, pp. 62-63. This is indeed the letter which JonathanLivengood cited in his comment above.

Comment author: JonathanLivengood 02 January 2013 06:32:18PM 2 points [-]

Wow! Thanks for the Good Thinking link. Now I won't have to scan it myself.

Comment author: JonathanLivengood 02 January 2013 06:30:11PM 1 point [-]

Yes, that's the letter!