Abstract
In this presentation, I discuss two features of robust Bayesian theory that arise, naturally, when considering more than one decision maker:
-
(1)
On a question of statics — what opportunities are there for Bayesians to engage in cooperative decision making while conforming the group to a (mild) Pareto principle and expected utility theory?
-
(2)
On a question of dynamics — what happens, particularly in the short run, when a collection of Bayesian opinions are updated using Bayes’ rule, where each opinion is conditioned on shared data?
In connection with the first problem — to allow for a Pareto efficient cooperative group — I argue for a relaxation of the “ordering” postulate in expected utility theory. I outline a theory of partially ordered preferences, developed in collaboration with Jay Kadane and Mark Schervish (1995), that relies on sets of probability / Utility pairs rather than a single such pair for making reobust Bayesian decisions.
In connection with the second problem — Tim Herron, Larry Wasserman, and I report on an anomalous phenomenon. We call it “dilation,” where conditioning on new shared evidence is certain to enlarge the range of opinions about an event of common interest. Dilation stands in contrast to well-known results about the asymptotic merging of Bayesian decision making to collect “cost-free” data prior to making a terminal decision?
The use of a set of probabilities to represent opinion, rather than the use of a single probability to do so, arises in the theory of random sets, e.g., when the random objects are events from a finite powerset. Then, as is well known, the set of probabilities is just that determined by the lower probability of a belief function. Dilation applies to belief functions, too, as I illustrate.
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
F.J. Anscombe and R.J. AumannA definition of subjective probabilityAnn. Math. Stat., 34 (1963), pp. 199–205.
J.O. BergerThe robust Bayesian viewpoint(with Discussion), Robustness of Bayesian Analysis (J.B. Kadane, ed.), Amsterdam. North—Holland, pp. 631–14,1984.
J.O. BergerStatistical Decision Theory(2na edition), Springer—Verlag, New York, 1985.
J.O. BergerRobust Bayesian analysis: Sensitivity to the priorJ. Stat. Planning and Inference, 25 (1990), pp. 303–328.
P. BlackAn examination of belief functions and other monotone capacitiesPh.D. Thesis, Department of Statistics, Carnegie Mellon University, 1996a.
P. BlackGeometric structure of lower probabilityThis Volume, pp. 361–383, 1996.
D. Blackwell and L. DubinsMerging of opinions with increasing informationAnn. Math. Stat., 33 (1962), pp. 882–887.
B. DefinettiLa prevision: Ses Lois logiques ses sources subjectivesAnnals de L’Institut Henri Poincarü, 7 (1937), pp. 1–68.
A.P. Dempster, Newmethods for reasoning towards posterior distributions based on sample dataAnn. Math. Stat., 37 (1966), pp. 355–374.
A.P. DempsterUpper and lower probabilities induced by a multi—valued mappingAnn. Math. Stat., 38 (1967), pp. 325–339.
D. EllsbergRisk ambiguity and the savage axiomsQuart. J. Econ., 75 (1961), pp. 643–669.
P.C. FishburnUtility Theory for Decision MakingKrieger Pub., New York, (1979 ed.), 1970.
P.C. FishburnLexicographic orders utilities and decision rules: A surveyManagement Science 20 (1974), pp. 1442–1471.
P.C. FishburnThe Axioms of subjective probabilityStatistical Science, 1 (1986), pp. 335–358.
P.C. Fishburn and I. LavalleSubjective expected lexicographic utility: Axioms and assessmentPreprint, AT&T Labs, Murray Hill, 1996.
I.J. GoodRational decisionsJ. Royal Stat. Soc., B14 (1952), pp. 107–114.
T. Herron, T. Seidenfeld, and L. WassermannDivisive conditionTechnical Report #585, Dept. of Statistics, Carnegie Mellon University, Pittsburgh, Pennsylvania, 1995.
K. Hestir, H. Nguyen, and G. RogersA random set formalism for evidential reasoningConditional Logic in Expert Systems, (I.R. Goodman, M.M. Gupta, H.T. Nguyen, and G.S. Rogers, eds.), Amsterdam: Elsevier Science, pp. 309–344, 1991.
P.J. HuberThe use of Choquet capacities in statisticsBull. Int. Stat., 45 (1973), pp. 181–191.
P.J. HuberRobust StatisticsWiley and Sons, New York, 1981.
P.J. Huber and V. StrassenMinimax tests and the Neyman-Pearson lemma for capacitiesAnnals of Statistics, 1 (1973), pp. 241–263.
R.C. JeffreyThe Logic of DecisionMcGraw Hill, New York, 1965.
H. JeffreysTheory of Probability3`d Edition, Oxford University Press, Oxford, 1967.
J.B. KadaneOpposition of interest in subjective Bayesian theoryManagement Science, 31 (1985), pp. 1586–1588.
H.E. KyburgProbability and Logic of Rational BeliefWesleyan University Press, Middleton, 1961.
H.E. KyburgBayesian and non-Bayesian evidential updatingArtificial Intelligence, 31 (1987), pp. 279–294.
M. LavineSensitivity in Bayesian statistics: The prior and the likelihoodJ. Amer. Stat. Assoc., 86 (1991), pp. 396–399.
I. LeviOn indeterminate probabilitiesJ. Phil., 71 (1974), pp. 391–418.
I. LeviConflict and social agencyJ. Phil., 79 (1982), pp. 231–247.
I. LeviImprecision and indeterminacy in probability judgmentPhil. of Science, 52 (1985), pp. 390–409.
I. LeviFeasibilityKnowledge, Belief, and Strategic Interaction, (C. Bicchieri and M.L. Dalla Chiara, eds.), Cambridge University Press, Cambridge, 1992.
M. MachinaExpected utility analysis without the independence axiomEconometrica, 50 (1982), pp. 277–323.
E.F. McclennenRationality and Dynamic ChoiceCambridge University Press, Cambridge, 1990.
L.R. Pericchi and P. WalleyRobust Bayesian credible intervals and prior ignoranceInt. Stat. Review, 58 (1991), pp. 1–23.
F.P. RamseyTruth and probabilityThe Foundations of Mathematics and Other Essays, Kegan, Paul, Trench Trubner and Co. Ltd., London, pp. 156–198, 1931.
L.J. SavageThe Foundations of StatisticsWiley and Sons, New York, 1954.
M.J. Schervish, T. Seidenfeld, and J.B. KadaneState-Dependent UtilitiesJ. Am. Stat. Assoc., 85 (1990), pp. 840–847.
M.J. Schervish, T. Seidenfeld, and J.B. KadaneShared preferences and state-dependent utilitiesManagement Science, 37 (1991), pp. 1575–1589.
T. SeidenfeldDecision theory without “independence” or without “ordering” - What is the difference?(with Discussion), Economics and Philosophy, 4 (1988), pp. 267–315.
T. Seidenfeld and M.J. SchervishConflict between finite additivity and avoiding Dutch bookPhil. of Science, 50 (1983), pp. 398–412.
T. Seidenfeld, M.J. Schervish, and J.B. KadaneDecisions without orderingActing and Reflecting, (W. Sieg, ed.), Kluwer Academic, Dordrecht, pp. 143–170, 1990.
T. Seidenfeld, M.J. Schervish, and J.B. KadaneA representation of partially ordered preferencesAnnals of Statistics, 23 (1995), pp. 2168–2217.
T. Seidenfeld and L. WassermanDilation for sets of probabilitiesAnnals of Statistics, 21 (1993), pp. 1139–1154.
C.A.B. SmithConsistency in statistical inference and decisionsJ. Royal Stat. Soc., B23 (1961), pp. 1–25.
E. SzpilrajnSur l’extension de l’ordre partielFund. Math., 16 (1930), pp. 386–389.
J. Von Neumann and O. MorgensternTheory of Games and Economic Behavior(2nd Edition), Princeton University Press, Princeton, New Jersey, 1947.
P. WalleyStatistical Reasoning with Imprecise ProbabilitiesChapman Hall, London, 1991.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer Science+Business Media New York
About this chapter
Cite this chapter
Seidenfeld, T. (1997). Some Static and Dynamic Aspects of Robust Bayesian Theory. In: Goutsias, J., Mahler, R.P.S., Nguyen, H.T. (eds) Random Sets. The IMA Volumes in Mathematics and its Applications, vol 97. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-1942-2_17
Download citation
DOI: https://doi.org/10.1007/978-1-4612-1942-2_17
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4612-7350-9
Online ISBN: 978-1-4612-1942-2
eBook Packages: Springer Book Archive