The Relation of Bayesian and Maximum Entropy Methods
Further progress in scientific inference must, in our view, come from some kind of unification of our present principles. As a prerequisite for this, we note briefly the great conceptual differences, and the equally great mathematical similarities, of Bayesian and Maximum Entropy methods.
KeywordsMaximum Entropy Method Exploratory Phase Hypothesis Space Maximum Entropy Principle Good Performance Feature
Unable to display preview. Download preview PDF.
- Bretthorst, L. (1987), Ph.D. Thesis, Department of Physics, Washington University, St. Louis, Missouri.Google Scholar
- Jaynes, E. T. (1968), “Prior Probabilities”, IEEE Trans. Systems Sci. Cybern. SSC-4, 227–241 (1968). Reprinted in V. M. Rao Tummala and R. C. Henshaw, Eds., Concepts and Applications of Modern Decision Methods (Michigan State University Business Studies Series, 1976), and in Jaynes (1983).Google Scholar
- Jaynes, E. T. (1983), Papers on Probability, Statistics and Statistical Physics”, R. D. Rosenkrantz, Editor, D. Reidel Pub. Co., Dordrecht-Holland. Reprints of 14 papers dated 1957-1980.Google Scholar
- Jeffrey, R. C. (1983), The Logic of Decision, 2nd Edition, Univ. of Chicago Press.Google Scholar
- van Campenhout, J. & Cover, T. M. (1981), IEEE Trans. Inform. Theory IT-27, 483.Google Scholar
- Zellner, A. (1987), “Optimal Information Processing and Bayes’ Theorem” The American Statistician (to be published).Google Scholar