Efficient Gibbs Sampling of Truncated Multivariate Normal with Application to Constrained Linear Regression, unpublished manuscript

In this paper we propose an efficient Gibbs sampler for simulation of a multivariate normal random vector subject to inequality linear constraints. Inference in a Bayesian linear model, where the regression parameters are subject to inequality linear constraints, is the primary motivation behind this research. In the literature, implementations of the Gibbs sampler for the multivariate normal distribution subject to inequality linear constraints and for the multiple linear regression with inequality constraints often exhibit poor mixing and slow convergence. This paper overcomes these limitations ∗ Gabriel Rodriguez-Yam is postdoctoral fellow and Richard Davis is Professor and chair