Local sensitivity of posterior expectations
Paul Gustafson
Source: Ann. Statist. Volume 24, Number 1 (1996), 174-195.
Abstract
We investigate the degree to which posterior expectations are sensitive to prior distributions, using a local method based on functional differentiation. Invariance considerations suggest a family of norms which can be used to measure perturbations to the prior. The sensitivity measure is seen to depend heavily on the choice of norm; asymptotic results suggest which norm will yield the most useful results in practice. We find that to maintain asymptotically sensible behaviour, it is necessary to reduce the richness of the class of prior perturbations as the dimension of the parameter space increases. Jeffreys' prior is characterized as the prior to which inference is least sensitive.
Primary Subjects: 62F15
Secondary Subjects: 62F35
Keywords: Classes of probabilities; Fréchet derivative; Jeffreys' prior; local sensitivity; $L^p$-norm; robustness
Full-text: Open access
Links and Identifiers
Permanent link to this document: http://projecteuclid.org/euclid.aos/1033066205
Mathematical Reviews number (MathSciNet):
MR1389886
Digital Object Identifier: doi:10.1214/aos/1033066205
Zentralblatt MATH identifier:
0853.62026
References
BASU, S., JAMMALAMADAKA, S. R. and LIU, W. 1993 . Local posterior robustness with parametric priors: maximum and average sensitivity. In Maximum Entropy and Bayesian MethZ . ods G. Heidbreder, ed. Kluwer, Dordecht. Z .
BERGER, J. O. 1986 . Comment on ``On the consistency of Bay es estimates'' by P. Diaconis and D. Freedman Ann. Statist. 14 30 37. Z .
BERGER, J. O. 1990 . Robust Bayesian analysis: sensitivity to the prior. J. Statist. Plann. Inference 25 303 328. Z .
BERGER, J. O. 1992 . Comment on ``Recent methodological advances in robust Bayesian inferZ ence'' by L. Wasserman. In Bayesian Statistics 4 J. M. Bernando, J. O. Berger, A. P. . Dawid, and A. F. M. Smith, eds. 495 496. Oxford Univ. Press. Z . Z .
BERGER, J. O. 1994 . An overview of robust Bayesian analysis with discussion . Test 3 5 124. Z .
Zentralblatt MATH:
0827.62026
BERGER, J. O. and BERLINER, L. M. 1986 . Robust Bay es and empirical Bay es analysis with -contaminated priors. Ann. Statist. 14 461 486. Z .
CHEN, C. F. 1985 . On asy mptotic normality of limiting density functions with Bayesian implications. J. Roy. Statist. Soc. Ser. B 47 540 546. Z .
CIFARELLI, D. M. and REGAZZINI, E. 1987 . On a general definition of concentration function. Sanky a Ser. B 49 307 319. Z .
Mathematical Reviews (MathSciNet):
MR91b:62026
Zentralblatt MATH:
0647.62023
CLARKE, B. S. and BARRON, A. R. 1994 . Jeffrey s' prior is asy mptotically least favourable under entropy risk. J. Statist. Plann. Inference 41 37 60. Z .
CUEVAS, A. and SANZ, P. 1988 . On differentiability properties of Bay es operators. In Bayesian Z . Statistics 3 J. M. Bernardo, M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds. 569 577. Oxford Univ. Press. Z .
Mathematical Reviews (MathSciNet):
MR90e:62006
DELAMPADY, M. and DEY, D. 1994 . Bayesian robustness for multiparameter problems. J. Statist. Plann. Inference 40 375 382. Z .
Mathematical Reviews (MathSciNet):
MR96e:62035
DEROBERTIS, L. 1978 . The use of partial prior knowledge in Bayesian inference. Ph.D. dissertation, Yale Univ.
DEROBERTIS, L. and HARTIGAN, J. 1981 . Bayesian inference using intervals of measures. Ann. Statist. 9 235 244. Z .
Mathematical Reviews (MathSciNet):
MR83e:62001
DEY, D. and BIRMIWAL, L. R. 1994 . Robust Bayesian analysis using entropy and divergence measures. Statist. Probab. Lett. 20 287 294. Z . Z .
Mathematical Reviews (MathSciNet):
MR95m:62054
DIACONIS, P. and FREEDMAN, D. 1986 . On the consistency of Bay es estimates with discussion . Ann. Statist. 14 1 67. Z .
Mathematical Reviews (MathSciNet):
MR88e:62016a
GELFAND, A. and DEY, D. 1991 . On Bayesian robustness of contaminated classes of priors. Statist. Decisions 9 63 80. Z . Z
Zentralblatt MATH:
0749.62007
GOOD, I. J. 1969 . What is the use of a distribution? In Multivariate Analy sis P. R. Krishnaiah, . ed. 2 183 203. Academic Press, New York. Z .
GUSTAFSON, P. 1996 . Local sensitivity of inferences to prior marginals. J. Amer. Statist. Assoc. 91 To appear. Z .
Mathematical Reviews (MathSciNet):
MR97f:62048
GUSTAFSON, P. and WASSERMAN, L. 1995 . Local sensitivity diagnostics for Bayesian inference. Ann. Statist. 23 2153 2167. Z .
HARTIGAN, J. A. 1983 . Bay es Theory. Springer, New York. Z .
Mathematical Reviews (MathSciNet):
MR85k:60008
JEFFREy S, H. 1961 . Theory of Probability, 3rd ed. Clarendon Press, Oxford. Z .
Mathematical Reviews (MathSciNet):
MR32:4710
KASS, R. E., TIERNEY, L. and KADANE, J. B. 1989 . Approximate methods for assessing influence and sensitivity in Bayesian analysis. Biometrika 76 663 674. Z .
Mathematical Reviews (MathSciNet):
MR91a:62083
Zentralblatt MATH:
0678.62037
LAVINE, M. 1991a . Sensitivity in Bayesian statistics: the prior and the likelihood. J. Amer. Statist. Assoc. 86 396 399. Z .
LAVINE, M. 1991b . An approach to robust Bayesian analysis for multidimensional parameter spaces. J. Amer. Statist. Assoc. 86 400 403. Z .
LINDLEY, D. V. 1992 . Comment on ``Recent methodological advances ion robust Bayesian Z inference'' by L. Wasserman. In Bayesian Statistics 4 J. M. Bernando, J. O. Berger, A. . P. Dawid and A. F. M. Smith, eds. 496. Oxford Univ. Press. Z .
RUGGERI, F. and WASSERMAN, L. 1993 . Infinitesimal sensitivity of posterior distributions. Canad. J. Statist. 21 195 203. Z .
Mathematical Reviews (MathSciNet):
MR94h:62045
RUGGERI, F. and WASSERMAN, L. 1995 . Density based classes of priors: infinitesimal properties and approximations. J. Statist. Plann. Inference 46 311 324. Z .
Mathematical Reviews (MathSciNet):
MR96k:62075
SIVAGANESAN, S. 1988 . Range of posterior measures for priors with arbitrary contaminations. Comm. Statist. Theory Methods 17 1591 1612. Z .
SIVAGANESAN, S. 1993 . Robust Bayesian diagnostics. J. Statist. Plann. Inference 35 171 188. Z .
Mathematical Reviews (MathSciNet):
MR95b:62003
SRINIVASAN, C. and TRUSZCZy NSKA, H. 1990 . On the ranges of posterior quantities. Technical Report 294, Dept. Statistics, Univ. Kentucky, Lexington. Z .
STEIN, C. 1985 . On the coverage probability of confidence sets based on a prior distribution. In Sequential Methods in Statistics 485 514. PWN Publishers, Warsaw. Z .
WASSERMAN, L. 1992a . Invariance properties of density ratio priors. Ann. Statist. 20 2177 2182. Z . Z
WASSERMAN, L. 1992b . Recent methodological advances in robust Bayesian inference with . Z discussion . In Bayesian Statistics 4 J. M. Bernando, J. O. Berger, A. P. Dawid and A. . F. M. Smith, eds. 483 502. Oxford Univ. Press. Z .
WASSERMAN, L. 1992c . The conflict between improper priors and robustness. Technical Report 559. Dept. Statistics, Carnegie Mellon Univ.
VANCOUVER, BRITISH COLUMBIA CANADA V6T 1Z2
The Annals of Statistics
