Imprecise Dirichlet Process With Application to the Hypothesis Test on the Probability That X ≤ Y

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.

[1]  Frank P. A. Coolen,et al.  A nonparametric predictive alternative to the Imprecise Dirichlet Model: The case of a known number of categories , 2009, Int. J. Approx. Reason..

[2]  Marco Zaffalon,et al.  A Bayesian Wilcoxon signed-rank test based on the Dirichlet process , 2014, ICML.

[3]  V. Susarla,et al.  On the posterior distribution of a dirichlet process given randomly right censored observations , 1977 .

[4]  Li Ma,et al.  Coupling Optional Pólya Trees and the Two Sample Problem , 2010, 1011.1253.

[5]  E. Lehmann,et al.  Nonparametrics: Statistical Methods Based on Ranks , 1976 .

[6]  S. Goodman.  Toward Evidence-Based Medical Statistics. 1: The P Value Fallacy , 1999, Annals of Internal Medicine.

[7]  E. Lehmann.  Elements of large-sample theory , 1998 .

[8]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

[9]  A. Raftery.  Bayesian Model Selection in Social Research , 1995 .

[10]  Peter Müller,et al.  DPpackage: Bayesian Semi- and Nonparametric Modeling in R , 2011 .

[11]  J. V. Ryzin,et al.  Nonparametric Bayesian Estimation of Survival Curves from Incomplete Observations , 1976 .

[12]  P. Walley,et al.  Robust Bayesian credible intervals and prior ignorance , 1991 .

[13]  Albert Y. Lo,et al.  A large sample study of the Bayesian bootstrap , 1987 .

[14]  A. Dasgupta.  Asymptotic Theory of Statistics and Probability , 2008 .

[15]  Raul Cano.  On The Bayesian Bootstrap , 1992 .

[16]  S. R. Dalai,et al.  Nonparametric Bayes Inference for Concordance in Bivariate Distributions , 1983 .

[17]  D. Blei.  Bayesian Nonparametrics I , 2016 .

[18]  D. Rubin.  The Bayesian Bootstrap , 1981 .

[19]  C. Holmes,et al.  Two-sample Bayesian Nonparametric Hypothesis Testing , 2009, 0910.5060.

[20]  T. Ferguson.  A Bayesian Analysis of Some Nonparametric Problems , 1973 .

[21]  Jean-Marc Bernard,et al.  An introduction to the imprecise Dirichlet model for multinomial data , 2005, Int. J. Approx. Reason..

[22]  John K Kruschke,et al.  Bayesian data analysis. , 2010, Wiley interdisciplinary reviews. Cognitive science.

[23]  E. Phadia.  Inference Based on Complete Data , 2013 .

[24]  J. L. Hodges,et al.  Estimates of Location Based on Rank Tests , 1963 .

[25]  Yuhui Chen,et al.  Bayesian nonparametric k-sample tests for censored and uncensored data , 2014, Comput. Stat. Data Anal..

[26]  Thomas Augustin,et al.  Nonparametric predictive inference and interval probability , 2004 .

[27]  M. Fay,et al.  Wilcoxon-Mann-Whitney or t-test? On assumptions for hypothesis tests and multiple interpretations of decision rules. , 2010, Statistics surveys.

[28]  Zoubin Ghahramani,et al.  Bayesian two-sample tests , 2009, ArXiv.

[29]  C. Weng,et al.  On a Second-Order Asymptotic Property of the Bayesian Bootstrap Mean , 1989 .

[30]  J. Sethuraman.  A CONSTRUCTIVE DEFINITION OF DIRICHLET PRIORS , 1991 .

[31]  J. Sethuraman,et al.  Convergence of Dirichlet Measures and the Interpretation of Their Parameter. , 1981 .

[32]  P. Walley.  Inferences from Multinomial Data: Learning About a Bag of Marbles , 1996 .

[33]  Ryan Martin,et al.  A nonparametric empirical Bayes framework for large-scale multiple testing. , 2011, Biostatistics.

[34]  E. Wagenmakers,et al.  Toward evidence‐based medical statistics: a Bayesian analysis of double‐blind placebo‐controlled antidepressant trials in the treatment of anxiety disorders , 2016, International journal of methods in psychiatric research.

[35]  Gert de Cooman,et al.  Representation insensitivity in immediate prediction under exchangeability , 2009, Int. J. Approx. Reason..

[36]  Marco Zaffalon,et al.  A Model of Prior Ignorance for Inferences in the One-parameter Exponential Family , 2012 .

[37]  James O. Berger,et al.  An overview of robust Bayesian analysis , 1994 .

[38]  F. Coolen,et al.  Nonparametric Predictive Inference for Reproducibility of Basic Nonparametric Tests , 2014 .

[39]  P. Walley.  Statistical Reasoning with Imprecise Probabilities , 1990 .

[40]  Nonparametric Bayesian robustness , 2010 .