Abstract
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(X ≤ Y). 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(X ≤ Y). 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.
Similar content being viewed by others
References
Augustin, T., and F. Coolen. 2004. Nonparametric predictive inference and interval probability. J. Stat. Plan. Inference, 124(2), 251–272.
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.
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.
Berger, J. O. 1994. An overview of robust Bayesian analysis with discussion. Test, 3(1), 5–124.
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.
Bernard, J.-M. 2005. An introduction to the imprecise Dirichlet model for multinomial data. Int. J. Approx. Reasoning, 39(23), 123–150.
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: http://dx.doi.org/10.1016/0304-4149(77)90030-8.
Borgwardt, K. M., and Z. Ghahramani. n.d. Bayesian two-sample tests. arXiv preprint arXiv:0906.4032.
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.
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.
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.
de Cooman, G., E. Miranda, and E. Quaeghebeur. 2009. Representation insensitivity in immediate prediction under exchangeability. Int. J. Approx. Reasoning, 50(2), 204–216.
Dalal, S., and E. Phadia. 1983. Nonparametric Bayes inference for concordance in bivariate distributions. Commun. Stat. Theory Methods, 12(8), 947–963.
DasGupta, A. 2008. Asymptotic theory of statistics and probability. New York, NY: Springer.
Efron, B. 1979. Bootstrap methods: Another look at the jackknife. Ann. Stat., 7(1), 1–26.
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.
Ferguson, T. S. 1973. A Bayesian analysis of some nonparametric problems. Ann. Stat., 1(2), 209–230.
Ghosh, J. K., and R. Ramamoorthi. 2003. Bayesian nonparametrics. New York, NY: Springer.
Goodman, S. N. 1999. Toward evidence-based medical statistics. 1: The P-value fallacy. Ann. Internal med., 130(12), 995–1004.
Hodges, J. L., and E. L. Lehmann. 1963. Estimates of location based on rank tests. Ann. Math. Stat., 34(2), 598–611.
Holmes, C., F. Caron, J. Griffin, and D. A. Stephens. n.d. Two-sample Bayesian nonparametric hypothesis testing. arXiv preprint arXiv:0910.5060.
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.
Kruschke, J. K. 2010. Bayesian data analysis. Wiley Interdiscip. Rev. Cognitive Sci., 1(5), 658–676.
Lehmann, E. 1998. Elements of large-sample theory, New York, NY: Springer-Verlag.
Lehmann, E., and H. J. M. D’Abrera. 1975. Nonparametrics: Statistical methods based on ranks. San Francisco, CA: McGraw-Hill.
Lo, A. Y. 1987. A large sample study of the Bayesian bootstrap. Ann. Stat., 15(1), 360–375.
Ma, L., and W. H. Wong. 2010. Coupling optional Pólya trees and the two sample problem, J. Am. Stat. Assoc., 106(496), 1553–1565.
Martin, R., and S. Tokdar. 2012. A nonparametric empirical Bayes framework for large-scale multiple testing. Biostatistics, 13(3), 427–439.
Pericchi, L. R., and P. Walley. 1991. Robust Bayesian credible intervals and prior ignorance. Int. Stat. Rev., 58(1), 1–23.
Phadia, E. G. 2013. Inference based on complete data. In Prior processes and their applications, 109–153. New York, NY: Springer.
Raftery, A. E. 1995. Bayesian model selection in social research. Sociol. Methodol., 25, 111–164.
Rossi, P., and R. McCulloch. 2010. Bayesm: Bayesian inference for marketing/micro-econometrics, R package version 2-2. http://www.perossi.org/home/bsm-1
Rubin, D. B. 1981. Bayesian bootstrap. Ann. Stat., 9(1), 130–134.
Ruggeri, F. 2010. Nonparametric Bayesian robustness. Chilean J. Stat., 2, 51–68.
Sethuraman, J. 1994. A constructive definition of Dirichlet priors. Stat. Sin., 4(2), 639–650.
Sethuraman, J., and R. C. Tiwari. 1981. Convergence of Dirichlet measures and the interpretation of their parameter. Defense Technical Information Center.
Susarla, V., and J. Van Ryzin. 1976. Nonparametric Bayesian estimation of survival curves from incomplete observations. J. Am. Stat. Assoc., 71(356), 897–902.
Walley, P. 1991. Statistical reasoning with imprecise probabilities. New York, NY: Chapman and Hall.
Walley, P. 1996. Inferences from multinomial data: Learning about a bag of marbles. J. R. Stat. Society. Ser. B (Methodological), 58(1), 3–57.
Weng, C.-S. 1989. On a second-order asymptotic property of the Bayesian bootstrap mean. Ann. Stat., 17(2), 705–710.
Author information
Authors and Affiliations
Corresponding author
Additional information
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ujsp.
Rights and permissions
About this article
Cite this article
Benavoli, A., Mangili, F., Ruggeri, F. et al. Imprecise Dirichlet Process With Application to the Hypothesis Test on the Probability That X ≤ Y . J Stat Theory Pract 9, 658–684 (2015). https://doi.org/10.1080/15598608.2014.985997
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1080/15598608.2014.985997