Updating Probabilities with Data and Moments

Abstract

We use the method of Maximum (relative) Entropy to process information in the form of observed data and moment constraints. The generic "canonical" form of the posterior distribution for the problem of simultaneous updating with data and moments is obtained. We discuss the general problem of non-commuting constraints, when they should be processed sequentially and when simultaneously. As an illustration, the multinomial example of die tosses is solved in detail for two superficially similar but actually very different problems.

arXiv commentPresented at the 27th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, Saratoga Springs, NY, July 8-13, 2007. 10 pages, 1 figure V2 has a small typo in the end of the appendix that was fixed. aj=mj+1 is now aj=m(k-j)+1

Cited by in corpus (11)

• Entropy, Information, and the Updating of Probabilities
• State retrieval beyond Bayes' retrodiction
• Application of the Maximum relative Entropy method to the physics of ferromagnetic materials
• Theoretical investigations of an information geometric approach to complexity
• Learning on dynamic statistical manifolds
• Bayesian Dropout
• On Entropy Regularized Path Integral Control for Trajectory Optimization
• Information Geometric Complexity of Entropic Motion on Curved Statistical Manifolds under Different Metrizations of Probability Spaces
• Complexity Characterization in a Probabilistic Approach to Dynamical Systems Through Information Geometry and Inductive Inference
• The Design of Global Correlation Quantifiers and Continuous Notions of Statistical Sufficiency
• Entropic Dynamics: A hybrid-contextual theory of Quantum Mechanics