by Zhongxue Chen and Monnie McGee

Journal of Data Science, v.6, no.2, 261-268

Abstract

The rule of three gives 3/n as the upper 95% bound for the success rate of the zero-numerator problems. However, this bound is usually conservative although it is useful in practice. Some Bayesian methods with beta distributions as priors have been studied. However, choosing the parameters for the priors is subjective and can severely impact the corresponding posterior distributions. In this paper, some hierarchical models are proposed, which provide practitioners other options for those zero-numerator problems.

Homepage | Table of Contents | Full Text of This Article