|
|
Re: Bayesian Book Recommendation
Posted:
Feb 13, 1999 8:32 AM
|
|
In article <36C02FE2.F3671E0@talkingxyzzy.com>,
Terry Braun <tab@talkingxyzzy.com> wrote:
> I am not a mathematician, nor a statistician, but I'm interested in
> understanding Bayesian statistics from a conceptual point of view.
> Can someone make some recommendations? I want to understand
> enough in order to think about decision making from a subjectivist
> perspective.
>
[Here talking as an 'veteran' (!) of nine months studying Bayesian inference,
so weight my opinion accordingly]. And I speak as an engineer whose maths
is/was pass degree level -- thus I can understand most statistics texts, but
the really strongly mathematical ones give me trouble.
It might be worth looking at David MacKay's Bayesian methods for neural
networks FAQ:
http://wol.ra.phy.cam.ac.uk/mackay/Bayes_FAQ.html
That will open up a whole vista -- and not just about neural nets. There is a
recommended reading list somewhere in that. Many of the books recommended by
previous posters are there. Also the book that suited my temperment &
background:
@Book{sivia,
author = "D.S. Sivia",
title = "Data Analysis, A Bayesian Tutorial",
publisher = "Oxford University Press",
year = "1996"
}
The previous URL will point you to Radford Neal's section on Bayesian learning
in the comp.ai.neural.nets FAQ. I like a lot Neal's recent book:
@Book{neal-bayesian-nn,
author = "R.M. Neal",
title = "Bayesian Learning for Neural Networks",
publisher = "Springer Verlag",
year = "1996"
}
Next call might be Larry Bretthorst's "Obituary of Edwin Thompson Jaynes July
5, 1922 - April 30, 1998": http://bayes.wustl.edu/etj/etj.html
Look there for many online (PostScript) papers by Jaynes. Initially, look for
"Probability as Logic" (the short paper, Dartmouth Workshop) and "Bayesian
Methods: General Background" (paper, Calgary workshop).
When you read those, you'll probably want to download everything by Jaynes,
including his unfinished textbook "Probability Theory: The Logic of Science",
(the latest version I _think_) available from:
http://omega.math.albany.edu:8008/JaynesBook.html
And you may be interested in a printed collection of Jaynes' papers:
@Book{rosenkrantz-jaynes,
author = "R.D. Rosenkrantz (ed.)",
title = "E.T. Jaynes: Papers on Probability, Statistics and
Statistical Physics",
publisher = "Kluwer",
year = "1983"
}
Bretthorst's own book "Bayesian Spectrum Analysis and Parameter Estimation",
Springer 1988, is now out of print and downloadable from somewhere on the
bayes.wustl site.
MacKay has the draft of a text downloadable from the site mentioned earlier.
Maybe the best of all introductions to Bayesian Inference is Tom Loredo's
"From Laplace to Supernova SN 1987A: Bayesian Inference in Astrophysics",
downloadable from:
http://astrosun.tn.cornell.edu/staff/loredo/bayes/tjl.html
Comparisons with the 'classical' method are given in: V. Barnett, Comparative
Statistical Inference, Wiley, 1982 (doubtful if in print).
And you probably cannot ignore:
@Book{jeffreys-pr,
author = "H. Jeffreys",
title = "Theory of Probability, 3rd ed.",
publisher = "Oxford University Press/Oxford Classics Series",
year = "1961/1998"
}
I may have forgotten something, so maybe look at:
http://www.infm.ulst.ac.uk/~jgc/book/bayes.html
and try some of the really comprehensive Bayesian Inference sites linked from
there.
Hope this helps,
Jon Campbell
Jonathan G Campbell Univ. Ulster Magee College Derry BT48 7JL N. Ireland
+44 1504 375367 JG.Campbell@ulst.ac.uk http://www.infm.ulst.ac.uk/~jgc/
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
|
|
