Bayesian Sampling Algorithms for the Sample Selection and Two-Part Models

This paper considers two models to deal with an outcome variable that contains a large fraction of zeros, such as individual expenditures on health care: a sample-selection model and a two-part model. The sample-selection model uses two possibly correlated processes to determine the outcome: a decision process and an outcome process; conditional on a favorable decision, the outcome is observed. The two-part model comprises uncorrelated decision and outcome processes. The paper addresses the issue of selecting between these two models. With a Gaussian specification of the likelihood, the models are nested and inference can focus on the correlation coefficient. Using a fully parametric Bayesian approach, I present sampling algorithms for the model parameters that are based on data augmentation. In addition to the sampler output of the correlation coefficient, a Bayes factor can be computed to distinguish between the models. The paper illustrates the methods and their potential pitfalls using simulated data sets