Bayesian Inference for Hidden Truncation Pareto (II) Models

摘要

Data from a hidden truncation Pareto (II) distribution are to be used to make inferences about the inequality, precision and truncation parameters. Two different types of dependent prior analyses are reviewed and compared with each other. It is argued that mathematical tractability should be, perhaps, a minor consideration in choosing the appropriate choices of the hyperparameters for the prior densities. Some illustrative examples are provided.