Bayesian hypothesis testing: Redux

Abstract

Bayesian hypothesis testing is re-examined from the perspective of an aprioriassessment of the test statistic distribution under the alternative. By assessing the distribution of an observable test statistic, rather than prior parameter values, we revisit the seminal paper of Edwards, Lindman and Savage (Psychol. Rev. 70 (1963) 193–242). There are a number of important take-aways from comparing the Bayesian paradigm via Bayes factors to frequentist ones. We provide examples where evidence for a Bayesian strikingly supports the null, but leads to rejection under a classical test. Finally, we conclude with directions for future research.