Abstract
The Minimum Cross-Entropy (MXE) inference rule leads to information systems which are inconsistent, and which may have an expectation less than the prior. The Min-Score rule (a generalization of maximum entropy) applied to information systems generates consistent systems and has a guaranteed expectation at least as great as the prior. The guaranteed expectation for the Min-Score rule is always at least as great as that for MXE.
Research supported in part by National Science Foundation grant IST 84-05161
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Dalkey, N. C. (1983). Updating Inductive Inference, presented at the third Workshop on Maximum Entropy and Bayesian Methods in Applied Statistics, University of Wyoming, August 1–4.
Dalkey, N. C. (1985). Inductive Inference and the Maximum Entropy Principle. In Maximum Entropy and Bayesian Methods in Inverse Problems, eds. C. Ray Smith and W. T. Grandy, pp. 351–64. Dordrect/Boston/Lancaster: D. Reidel.
Dalkey, N. C. (1986). Prior Probabilities Revisited. In Maximum Entropy and Bayesian Methods in Applied Statistics, ed. J. H. Justice, pp. 117–30. Cambridge: Cambridge University Press.
Shore, J. E. & Johnson, R. W. (1981) Properties of Cross-Entropy Minimization. IEEE Transactions on Information Theory, IT-27, 472–82.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1988 Kluwer Academic Publishers
About this chapter
Cite this chapter
Dalkey, N.C. (1988). Comparison of Minimum Cross-Entropy Inference with Minimally Informative Information Systems. In: Erickson, G.J., Smith, C.R. (eds) Maximum-Entropy and Bayesian Methods in Science and Engineering. Fundamental Theories of Physics, vol 31-32. Springer, Dordrecht. https://doi.org/10.1007/978-94-009-3049-0_18
Download citation
DOI: https://doi.org/10.1007/978-94-009-3049-0_18
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-7871-9
Online ISBN: 978-94-009-3049-0
eBook Packages: Springer Book Archive