Bayesian Analysis of the Stochastic Switching Regression Model Using Markov Chain Monte Carlo Methods

This study develops Bayesian methods for estimating the parameters of astochastic switching regression model. Markov Chain Monte Carlo methods, dataaugmentation, and Gibbs sampling are used to facilitate estimation of theposterior means. The main feature of these methods is that the posterior meansare estimated by the ergodic averages of samples drawn from conditionaldistributions, which are relatively simple in form and more feasible to samplefrom than the complex joint posterior distribution. A simulation study isconducted comparing model estimates obtained using data augmentation, Gibbssampling, and the maximum likelihood EM algorithm and determining the effectsof the accuracy of and bias of the researcher's prior distributions on theparameter estimates.