The Prior Predictive Value
At the heart of Bayesian model comparison lies the so-called prior-predictive value. In the important class of Quantified-Maxent applications analytic approximations are routinely used which often give rise to noise-fitting and ringing. We present an improved analytic expression which overcomes these short comings. In most interesting real-world problems, however, a standard approximation and straight forward application of Markov-Chain Monte Carlo are hampered by the complicated structure of the liklihood in parameter space. At the Maxent workshop 1997 in Boise John Skilling suggested to employ a formalism, borrowed from statistical physics, to compute the prior predictive value.
Bibliographic Reference: Article: Maximum Entropy and Bayesian Methods, (1999), 319-326
Record Number: 199911401 / Last updated on: 1999-10-01
Original language: en
Available languages: en