Abstract
Range of the posterior probability of an interval over the ε-contamination class Γ={π=(1−ε)π0+εq:qεQ} is derived. Here, π0 is the elicited prior which is assumed unimodal, ε is the amount of uncertainty in π0, andQ is the set of all probability densitiesq for which π=(1−ε)π0+εq is unimodal with the same mode as that of π0. We show that the sup (resp. inf) of the posterior probability of an interval is attained by a prior which is equal to (1−ε)π0 except in one interval (resp. two disjoint intervals) where it is constant.
Similar content being viewed by others
References
Basu, S. and DasGupta, A. (1990). Bayesian analysis under distribution bands, Tech. Report 90-48, Department of Statistics, Purdue University, Indiana.
Berger, J. (1984). The robust Bayesian viewpoint (with discussion),Robustness of Bayesian Analysis (ed. J. Kadane), 63–144, North-Holland, Amsterdam.
Berger, J. (1985).Statistical Decision Theory and Bayesian Analysis, Springer, New York.
Berger, J. (1990). Robust Bayesian analysis: sensitivity to the prior,J. Statist. Plann. Inference,25, 303–328.
Berger, J. and Berliner, L. M. (1986). Robust Bayes and empirical Bayes analysis with ε-contaminated priors,Ann. Statist.,14, 461–486.
Berger, J. and O'Hagan, A. (1988). Ranges of posterior probabilities for unimodal priors with specified quantiles,Bayesian Statistics III (eds. J. M. Bernardo, M. H. Degroot, D. V. Lindley, and A. F. M. Smith), Oxford University Press.
Berliner, L. M. and Goel, P. (1986). Incorporating partial prior information: range of posterior probabilities, Tech. Report 357, Department of Statistics, Ohio State University.
Bose, S. (1990). Bayesian robustness with shape constrained priors and mixture priors, Ph.D. Thesis, Purdue University, Indiana.
DasGupta, A. (1991). Diameter and volume minimizing confidence sets in Bayes and classical problems,Ann. Statist.,19, 1225–1243.
DasGupta, A. and Studden, W. (1991). Robust Bayesian experimental designs in normal linear models,Ann. Statist.,19, 1244–1256.
DeRobertis, L. and Hartigan, J. A. (1981). Bayesian inference using intervals of measures,Ann. Statist.,9, 235–244.
Good, I. J. (1983).Good Thinking: The Foundations of Probability and Its Applications, University of Minnesota Press, Minneapolis.
Huber, P. J. (1973). The use of Choquet capacities in statistics,Proc. 39th Session ISI,45, 181–188.
Lavine, M. (1991a). Sensitivity in Bayesian statistics: the prior and likelihood,J. Amer. Statist. Assoc.,86, 396–399.
Lavine, M. (1991b). An approach to robust Bayesian analysis with multidimensional problems,J. Amer. Statist. Assoc.,86, 400–403.
Lavine, M., Wasserman, L. and Wolpert, R. (1991). Bayesian inference with specified prior marginals,J. Amer. Statist. Assoc.,86, 964–971.
Moreno, E. and Cano, J. A. (1988). Robust Bayesian analysis with ε-contaminations partially known, Tech. Report, University of Granada, Spain.
O'Hagan, A. and Berger, J. (1988). Ranges of posterior probabilities for quasiunimodal priors with specified quantiles,J. Amer. Statist. Assoc.,83, 503–508.
Polasek, W. (1985). Sensitivity analysis for general and hierarchical linear regression models,Bayesian Inference and Decision Techniques with Applications (eds. P. K. Goel and A. Zellner), North-Holland, Amsterdam.
Sivaganesan, S. (1989). Sensitivity of posterior mean to unimodality preserving contaminations,Statist. Decisions,7, 77–93.
Sivaganesan, S. (1991). Sensitivity of some posterior summaries when prior is unimodal with specified quantiles,Canad. J. Statist.,19, 57–66.
Sivaganesan, S. and Berger, J. (1989). Ranges of posterior measures for priors with unimodal contaminations,Ann. Statist.,17, 868–889.
Walley, P. (1990).Statistical Reasoning with Imprecise Probabilities, Chapman and Hall, London.
Author information
Authors and Affiliations
About this article
Cite this article
Sivaganesan, S. Range of the posterior probability of an interval for priors with unimodality preserving contaminations. Ann Inst Stat Math 45, 187–199 (1993). https://doi.org/10.1007/BF00773678
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF00773678