Summary
The formulation of the concept of non-informative prior distribution over a finite number of possibilities is considered and the minimum information prior distribution is defined as the prior distribution that adds minimum expected amount of information to the posterior distribution. Numerical examples show that the definition leads to nontrivial results. An information inequality is established to assure the validity of numerical results. The relation of the present work to other works on the same subject is briefly reviewed and finally a minimax type prior distribution is introduced that exhibits the impartial property which is lacking in the minimum information prior distribution.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Akaike, H.. (1977). On entropy maximization principl,tApplications of Statistics (ed. P. R. Krishnaiah) North-Holland, Amsterdam, 27–41.
Akaike, H.. (1983). Statistical inference and the measurement of entropy,Scientific Inference, Data Analysis and Robustness (eds. G.E.P. Box, T. Leonard and C. F. Wu), Academic Press, New York, 165–189.
Bernardo, J. M.. (1979). Reference posterior distributions for Bayesian inference (with discussion).J. R. Statist. Soc., B,41, 113–147.
Cox, D. R. and Hinkley, D. V.. (1974),Theoretical Statistics, Chapman & Hall, London.
Jeffreys, H.. (1946). An invariant form for the prior probability in estimation problems.Proceedings of the Royal Society of London, Ser. A,186, 453–461.
Kullback, S. and Leibler, R. A. (1951). On information and sufficiency,Ann. Math. Statist.,22, 79–86,
Lindley, D. V.. (1956). On a measure of the information provided by an experiment,Ann. Math. Statist.,27, 986–1005.
Zellner, A. (1977). Maximal data information prior distributions,New Developments in the Applications of Bayesian Methods (eds A. Aykac and C. Brumat), North-Holland, Amsterdam 211–232.
Additional information
This work was partly supported by the United States Army Contract No. DAAG 29-80-C-0041 in Mathematics Research Center, University of Wisconsin-Madison.
The Institute of statistical mathematics
About this article
Cite this article
Akaike, H. On minimum information prior distributions. Ann Inst Stat Math 35, 139–149 (1983). https://doi.org/10.1007/BF02480970
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF02480970