Journal of Statistical Theory and Practice

, Volume 9, Issue 3, pp 658–684 | Cite as

Imprecise Dirichlet Process With Application to the Hypothesis Test on the Probability That XY

  • Alessio BenavoliEmail author
  • Francesca Mangili
  • Fabrizio Ruggeri
  • Marco Zaffalon


The Dirichlet process (DP) is one of the most popular Bayesian nonparametric models. An open problem with the DP is how to choose its infinite-dimensional parameter (base measure) in case of lack of prior information. In this work, we present the imprecise DP (IDP)—a prior near-ignorance DP-based model that does not require any choice of this probability measure. It consists of a class of DPs obtained by letting the normalized base measure of the DP vary in the set of all probability measures. We discuss the tight connections of this approach with Bayesian robustness and in particular prior near-ignorance modeling via sets of probabilities. We use this model to perform a Bayesian hypothesis test on the probability P(XY). We study the theoretical properties of the IDP test (e.g., asymptotic consistency), and compare it with the frequentist Mann-Whitney-Wilcoxon rank test that is commonly employed as a test on P(XY). In particular, we show that our method is more robust, in the sense that it is able to isolate instances in which the aforementioned test is virtually guessing at random.


Bayesian nonparametric test Imprecise Dirichlet process Wilcoxon rank sum 

AMS Subject Classification

62G10 62G35 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Augustin, T., and F. Coolen. 2004. Nonparametric predictive inference and interval probability. J. Stat. Plan. Inference, 124(2), 251–272.MathSciNetCrossRefGoogle Scholar
  2. Benavoli, A., and M. Zaffalon. 2012. A model of prior ignorance for inferences in the one-parameter exponential family. J. Stat. Plan. Inference, 142(7), 1960–1979.MathSciNetCrossRefGoogle Scholar
  3. Benavoli, A., F. Mangili, G. Corani, M. Zaffalon, and F. Ruggeri. 2014. A Bayesian Wilcoxon signed-rank test based on the Dirichlet process. Proceedings of the 32th International Conference on Machine Learning (ICML), 1026–1034.Google Scholar
  4. Berger, J. O. 1994. An overview of robust Bayesian analysis with discussion. Test, 3(1), 5–124.MathSciNetCrossRefGoogle Scholar
  5. Berger, J. O., D. Rios Insua, and F. Ruggeri. 2000. Bayesian robustness. In Robust Bayesian analysis, ed. D. Rios Insua and F. Ruggeri, vol. 152 of Lecture Notes in Statistics, 1–32. New York, NY: Springer.CrossRefGoogle Scholar
  6. Bernard, J.-M. 2005. An introduction to the imprecise Dirichlet model for multinomial data. Int. J. Approx. Reasoning, 39(23), 123–150.MathSciNetCrossRefGoogle Scholar
  7. Blum, J., and V. Susarla. 1977. On the posterior distribution of a dirichlet process given randomly right censored observations. Stochastic Processes Their Appl., 5(3), 207–211. doi: Scholar
  8. Borgwardt, K. M., and Z. Ghahramani. n.d. Bayesian two-sample tests. arXiv preprint arXiv:0906.4032.Google Scholar
  9. Chen, Y., and T. E. Hanson. 2014. Bayesian nonparametric k-sample tests for censored and uncensored data. Comput. Stat. Data Anal., 71(C), 335–346.MathSciNetCrossRefGoogle Scholar
  10. Coolen, F., and T. Augustin. 2009. A nonparametric predictive alternative to the imprecise Dirichlet model: The case of a known number of categories. Int. J. Approximate Reasoning, 50(2), 217–230.MathSciNetCrossRefGoogle Scholar
  11. Coolen, F. P., and S. Bin Himd. 2014. Nonparametric predictive inference for reproducibility of basic nonparametric tests. J. Stat. Theory Pract., 8(4), 591–618.MathSciNetCrossRefGoogle Scholar
  12. de Cooman, G., E. Miranda, and E. Quaeghebeur. 2009. Representation insensitivity in immediate prediction under exchangeability. Int. J. Approx. Reasoning, 50(2), 204–216.MathSciNetCrossRefGoogle Scholar
  13. Dalal, S., and E. Phadia. 1983. Nonparametric Bayes inference for concordance in bivariate distributions. Commun. Stat. Theory Methods, 12(8), 947–963.MathSciNetCrossRefGoogle Scholar
  14. DasGupta, A. 2008. Asymptotic theory of statistics and probability. New York, NY: Springer.zbMATHGoogle Scholar
  15. Efron, B. 1979. Bootstrap methods: Another look at the jackknife. Ann. Stat., 7(1), 1–26.MathSciNetCrossRefGoogle Scholar
  16. Fay, M. P., and M. A. Proschan. 2010. Wilcoxon-Mann-Whitney or t-test? On assumptions for hypothesis tests and multiple interpretations of decision rules. Stat. Surveys, 4, 1–39.MathSciNetCrossRefGoogle Scholar
  17. Ferguson, T. S. 1973. A Bayesian analysis of some nonparametric problems. Ann. Stat., 1(2), 209–230.MathSciNetCrossRefGoogle Scholar
  18. Ghosh, J. K., and R. Ramamoorthi. 2003. Bayesian nonparametrics. New York, NY: Springer.zbMATHGoogle Scholar
  19. Goodman, S. N. 1999. Toward evidence-based medical statistics. 1: The P-value fallacy. Ann. Internal med., 130(12), 995–1004.CrossRefGoogle Scholar
  20. Hodges, J. L., and E. L. Lehmann. 1963. Estimates of location based on rank tests. Ann. Math. Stat., 34(2), 598–611.MathSciNetCrossRefGoogle Scholar
  21. Holmes, C., F. Caron, J. Griffin, and D. A. Stephens. n.d. Two-sample Bayesian nonparametric hypothesis testing. arXiv preprint arXiv:0910.5060.Google Scholar
  22. Jara, A., T. Hanson, F. Quintana, P. Müller, and G. Rosner. 2011. DPpackage: Bayesian semi- and nonparametric modeling in R. J. Stat. Software, 40(5), 1–30.CrossRefGoogle Scholar
  23. Kruschke, J. K. 2010. Bayesian data analysis. Wiley Interdiscip. Rev. Cognitive Sci., 1(5), 658–676.CrossRefGoogle Scholar
  24. Lehmann, E. 1998. Elements of large-sample theory, New York, NY: Springer-Verlag.Google Scholar
  25. Lehmann, E., and H. J. M. D’Abrera. 1975. Nonparametrics: Statistical methods based on ranks. San Francisco, CA: McGraw-Hill.zbMATHGoogle Scholar
  26. Lo, A. Y. 1987. A large sample study of the Bayesian bootstrap. Ann. Stat., 15(1), 360–375.MathSciNetCrossRefGoogle Scholar
  27. Ma, L., and W. H. Wong. 2010. Coupling optional Pólya trees and the two sample problem, J. Am. Stat. Assoc., 106(496), 1553–1565.CrossRefGoogle Scholar
  28. Martin, R., and S. Tokdar. 2012. A nonparametric empirical Bayes framework for large-scale multiple testing. Biostatistics, 13(3), 427–439.CrossRefGoogle Scholar
  29. Pericchi, L. R., and P. Walley. 1991. Robust Bayesian credible intervals and prior ignorance. Int. Stat. Rev., 58(1), 1–23.CrossRefGoogle Scholar
  30. Phadia, E. G. 2013. Inference based on complete data. In Prior processes and their applications, 109–153. New York, NY: Springer.CrossRefGoogle Scholar
  31. Raftery, A. E. 1995. Bayesian model selection in social research. Sociol. Methodol., 25, 111–164.CrossRefGoogle Scholar
  32. Rossi, P., and R. McCulloch. 2010. Bayesm: Bayesian inference for marketing/micro-econometrics, R package version 2-2.
  33. Rubin, D. B. 1981. Bayesian bootstrap. Ann. Stat., 9(1), 130–134.MathSciNetCrossRefGoogle Scholar
  34. Ruggeri, F. 2010. Nonparametric Bayesian robustness. Chilean J. Stat., 2, 51–68.MathSciNetzbMATHGoogle Scholar
  35. Sethuraman, J. 1994. A constructive definition of Dirichlet priors. Stat. Sin., 4(2), 639–650.MathSciNetzbMATHGoogle Scholar
  36. Sethuraman, J., and R. C. Tiwari. 1981. Convergence of Dirichlet measures and the interpretation of their parameter. Defense Technical Information Center.Google Scholar
  37. Susarla, V., and J. Van Ryzin. 1976. Nonparametric Bayesian estimation of survival curves from incomplete observations. J. Am. Stat. Assoc., 71(356), 897–902.MathSciNetCrossRefGoogle Scholar
  38. Walley, P. 1991. Statistical reasoning with imprecise probabilities. New York, NY: Chapman and Hall.CrossRefGoogle Scholar
  39. Walley, P. 1996. Inferences from multinomial data: Learning about a bag of marbles. J. R. Stat. Society. Ser. B (Methodological), 58(1), 3–57.MathSciNetzbMATHGoogle Scholar
  40. Weng, C.-S. 1989. On a second-order asymptotic property of the Bayesian bootstrap mean. Ann. Stat., 17(2), 705–710.MathSciNetCrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2015

Authors and Affiliations

  • Alessio Benavoli
    • 1
    Email author
  • Francesca Mangili
    • 1
  • Fabrizio Ruggeri
    • 2
  • Marco Zaffalon
    • 1
  1. 1.IPG IDSIAMannoSwitzerland
  2. 2.CNR IMATIMilanoItaly

Personalised recommendations