Block unimodality for multivariate Bayesian robustness

  • Brunero Liseo
  • Lea Petrella
  • Gabriella Salinetti


The development of Bayesian robustness has been growing in the last decade. The theory has extensively dealt with the univariate parameter case. Among the vast amount of proposals in the literature, only a few of them have a straightforward extension to the multivariate case. In this paper we consider the multidimensional version of the class of ε-contaminated prior distributions, with unimodal contaminations. In the multivariate case there is not a unique definition of unimodality and one's choice must be based on statistical ground. Here we propose the use of the block unimodal distributions, which proved to be very suitable for modelling situations where the coordinates of the parameter ϑ are deemed, a priori, weakly correlated.


Multivariate Unimodality Bayesian Inference Robustness ε-contaminated class of priors 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Berger, J. O. andBernardo, J. M. (1989). Estimating the product of means: Bayesian analysis with reference priors.Journal of the American Statistical Association 84, 200–207.zbMATHCrossRefMathSciNetGoogle Scholar
  2. Berger, J. O. andMoreno, E. (1993). Bayesian Robustness in bidimensional models: prior independence.Journal of Statistical Planning and Inference (to appear).Google Scholar
  3. Bose, S. (1990). Bayesian Robustness with Shape-Constrained Priors and Mixture Priors.Ph. D. Thesis, Dept. of Statistics, Purdue University.Google Scholar
  4. Cox, D. R. andHinkley, D. V. (1974).Theoretical Statistics, Chapman and Hall, London.zbMATHGoogle Scholar
  5. DasGupta, A., Ghosh, J. K. andZen, M. M. (1990). Bayesian analysis of a multivariate normal mean with flat tailed priors.Technical Report 90-22. Dept. of Statistics, Purdue University.Google Scholar
  6. Dharmadhikari, S. andJoag-Dev, K. (1988).Unimodality, Convexity and Applications, Academic Press, San Diego.zbMATHGoogle Scholar
  7. Huber, P. J. (1973).The use of Choquet capacities in statistics.Bull. Int. Statist. Inst. 45, 181–191.MathSciNetGoogle Scholar
  8. Kass, R. andGreenhouse, J. (1989). Investigating therapies of potentially great benefit: A Bayesian perspective. Comments on «Investigating therapies of potentially great benefit: ECMO», by J. H. Ware.Statistical Science 4, 310–317.Google Scholar
  9. Lavine, M. (1991). An approach to Robust Bayesian Analysis for Multidimensional Parameter Spaces.Journal of the American Statistical Association 86, 400–403.zbMATHCrossRefMathSciNetGoogle Scholar
  10. Lavine, M., Wasserman, L. andWolpert, R. (1991). Bayesian Inference with specified prior marginals.Journal of the American Statistical Association 86, 964–971.zbMATHCrossRefMathSciNetGoogle Scholar
  11. Shepp, L. A. (1962). Symmetric Random Walk.Trans. Amer. Math. Soc. 104, 144–153.zbMATHCrossRefMathSciNetGoogle Scholar
  12. Sivaganesan, S. andBerger, J. O. (1989). Ranges of posterior measures for priors with unimodal contaminations.Annals of Statistics 17, 868–889.zbMATHMathSciNetGoogle Scholar
  13. Ware, J. H. (1989). Investigating therapies of potentially great benefit: ECMO.Statistical Science 4, 310–317.MathSciNetGoogle Scholar
  14. Wassermann, L. (1992). Recent Methodological Advances in Robust Bayesian Inference.Bayesian Statistics 4, pp. 483–502 (J. M. Bernardo, J. O. Berger, A. P. Dawid, A. F. M. Smith Eds.). Oxford University Press.Google Scholar

Copyright information

© Società Italiana di Statistica 1993

Authors and Affiliations

  • Brunero Liseo
    • 3
  • Lea Petrella
    • 1
  • Gabriella Salinetti
    • 2
  1. 1.Dip. StatisticoUniversità degli Studi di FirenzeFirenzeItaly
  2. 2.Università di Roma «La Sapienza»RomaItaly
  3. 3.Dipartimento di Statistica, Probabilità e Statistiche ApplicateUniversità di Roma «La Sapienza»RomaItaly

Personalised recommendations