The Annals of Statistics

Asymptotic behavior of Bayes estimates under possibly incorrect models

Olaf Bunke and Xavier Milhaud

Full-text: Open access


We prove that the posterior distribution in a possibly incorrect parametric model a.s. concentrates in a strong sense on the set of pseudotrue parameters determined by the true distribution. As a consequence, we obtain in the case of a unique pseudotrue parameter the strong consistency of pseudo-Bayes estimators w.r.t. general loss functions.

Further, we present a simple example based on normal distributions and having two different pseudotrue parameters, where pseudo-Bayes estimators have an essentially different asymptotic behavior than the pseudomaximum likelihood estimator. While the MLE is strongly consistent, the sequence of posterior means is strongly inconsistent and a.s. almost all its accumulation points are not pseudotrue. Finally, we give conditions under which a pseudo-Bayes estimator for a unique pseudotrue parameter has an asymptotic normal distribution.

Article information

Ann. Statist., Volume 26, Number 2 (1998), 617-644.

First available in Project Euclid: 31 July 2002

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62F12: Asymptotic properties of estimators 62F15: Bayesian inference

Consistency asymptotic normality incorrect parametric models inconsistent Bayes estimates


Bunke, Olaf; Milhaud, Xavier. Asymptotic behavior of Bayes estimates under possibly incorrect models. Ann. Statist. 26 (1998), no. 2, 617--644. doi:10.1214/aos/1028144851.

Export citation


  • BERK, R. H. 1966. Limiting behaviour of posterior distributions when the model is incorrect. Ann. Math. Statist. 37 51 58. Z.
  • BERK, R. H. 1970. Consistency of a posteriori. Ann. Math. Statist. 41 894 906. Z.
  • BICKEL, P. J. and FREEDMAN, D. 1981. Some asy mptotic theory for the bootstrap. Ann. Statist. 9 1196 1217. Z.
  • BICKEL, P. J. and YAHAV, J. A. 1969. Some contributions to the asy mptotic theory of Bay es solutions. Z. Wahrsch. Verw. Gebiete 11 257 276. Z.
  • CHUNG, K. L. 1974. A Course in Probability Theory, 2nd ed. Academic Press, New York. Z. DACUNHA-CASTELLE, D. and DUFLO, M. 1983. Probabilites et Statistiques 2. Problemes a Temps ´ Mobile. Masson, Paris. Z.
  • DIACONIS, P. and FREEDMAN, D. 1986a. On the consistency of Bay es estimates. Ann. Statist. 14 1 26. Z.
  • DIACONIS, P. and FREEDMAN, D. 1986b. On inconsistent Bay es estimates of location. Ann. Statist. 14 68 87. Z.
  • GOURIEROUX, C. and MONTFORT, A. 1993. Pseudo likelihood methods. In Handbook of Statistics Z. S. Kotz and N. L. Johnson, eds. 11, Chapter 12. North-Holland, Amsterdam. Z.
  • HALL, P. and HEy DE, C. C. 1980. Martingale Limit Theory and Its Application. Academic Press, New York. Z. HANOUSEK and JURECKOVA, J. 1996. Unpublished manuscript. Z.
  • HUBER, P. 1967. The behavior of maximum likelihood estimates under nonstandard conditions. Proc. Fifth Berkeley Sy mp. Math. Statist. Probab. 1 221 233. Univ. California Press, Berkeley. Z.
  • IBRAGIMOV, I. A. and HAS'MINSKII, R. Z. 1981. Statistical Estimation. Asy mptotic Theory. Springer, New York. Z.
  • LEDOUX, M. and TALAGRAND, M. 1991. Probability in Banach Spaces. Springer, Berlin. Z.
  • LEHMANN, E. L. 1983. Theory of Point Estimation. Wiley, New York. Z.
  • PFANZAGL, J. 1969. On the measurability and consistency of minimum contrast estimators. Metrika 14 249 272. Z.
  • STRASSER, H. 1981. Consistency of maximum likelihood and Bay es estimates. Ann. Statist. 9 1107 1113.