On Sample-Size Selection and the Evaluation of Discriminability in the Model Choice Problem
提出一种贝叶斯方法，通过两个二项模型间的差异来评估不同分布族（如von Mises与包裹正态分布）的可区分性，并探讨样本量选择的影响。
Abstract In this article we present a general Bayesian approach to discrimination. We describe how the difference between two separate families of distributions can be assessed in terms of differences between two binomial models. Two examples that illustrate the methodology are included. Specifically, we investigate the extent to which it is possible to discriminate between the von Mises and the wrapped normal distributions and also between the normal and a family of t distributions. Key Words: Bayesian inferenceDiscriminationBinomial familyVon Mises distributionWrapped normal distribution