Bayesian analysis of the SUR Tobit model

Abstract

This paper develops a practical sampling scheme for Bayesian analysis of correlated censored data using the seemingly unrelated Tobit regressions model. Posterior inference is performed via the Gibbs sampler with data augmentation algorithm. In particular, the relevant full conditional distributions needed in the use of Gibbs sampler are derived. The method is then applied to a real data set on the determination of the payments of cash and stock dividends.