Abstract
Probabilistic methods to make inference from a random sample of observations should satisfy certain requirements for consistency. A principle of invariance under conditioning of the random variable domain is stated. Such a principle is formalized using the paradigm of a two-step inference method based on data. Invariance requirements are expressed both in the data encoding and probability distribution selection steps.
Distributions may be assigned by minimization of cross-entropy using fractile constraints encoding data and any reference distribution. The paper shows that this version of entropy method satisfies the data encoding and probability selection invariance requirements.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
JEFFREYS, H. (1939, 2nd and 3th ed. 1948, 1961) Theory of probability, Oxford University Press, London.
JAYNES, E.T. (1957) “How Does the Brain Do Plausible Reasoning?”, Microwave Laboratory Report No. 421. University of Stanford; and G.J. Erickson and C.R. Smith (eds.). Maximum-Entropy and Bayesian Methods in Science and Engineering. Vol. I, 1–24, Kluwer Academic Publications, 1988.
SHORE J.E. and JOHNSON, R.W. (1980) “Axiomatic Derivation of Principle of Maximum Entropy and the Principle of Minimum Cross-Entropy” IEEE Transactions on Information Theory, IT-26, 1, 26–37.
TIKOCHINSKY, Y., TISHBY, N.Z. and LEVINE R.D. (1984) “Consistent Inference Probabilities for Reproducible Experiments”, Physics Review Letters, V. 51, 1357–1360.
SKILLING, J. (1988) “The axioms of Maximum Entropy”, G.J. Erickson and C.R. Smith (eds.). Maximum-Entropy and Bayesian Methods in Science and Engineering, Vol. 1, 173–187, Kluwer Academic Publications.
SOLANA, V. and LIND, N.C. (1989) “A Monotonic Property of Distributions Based on Entropy with Fractile Constraints”, 8th. International Maximum Entropy Conference, St. John’s College, Cambridge, England 1988; John Skilling (ed.), Kluwer Academic Publications, Dordrecht, Netherland.
SOLANA, V. AND LIND, N.C. (1989) “Two Principles for Data Based Probabilistic System Analysis”, Proceedings 5th. International Conference on Structural Safety and Reliability, San Francisco, August 1989. Shinozuka (ed.), ASCE American Society of Civil Engineers.
LIND, N.C. and SOLANA, V. (1988) “Estimation of Random Variables with Fractile Constraints”, IRR No. 11, Institute for Risk Research, University of Waterloo, Waterloo, Ontario, Canada.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1990 Kluwer Academic Publishers
About this chapter
Cite this chapter
Solana, V. (1990). Consistency Principle for Data-Based Probabilistic Inference. In: Fougère, P.F. (eds) Maximum Entropy and Bayesian Methods. Fundamental Theories of Physics, vol 39. Springer, Dordrecht. https://doi.org/10.1007/978-94-009-0683-9_4
Download citation
DOI: https://doi.org/10.1007/978-94-009-0683-9_4
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-6792-8
Online ISBN: 978-94-009-0683-9
eBook Packages: Springer Book Archive