Hidden Markov Model Regression

Hidden Markov Model Regression (HMMR) is an extension of the Hidden Markov Model (HMM) to regression analysis. We assume that the parameters of the regression model are determined by the outcome of a nite-state Markov chain and that the error terms are conditionally independent normally distributed with mean zero and state dependent variance. The theory of HMM regression is quite new, but some of its development calls on the natural extension of the work by Baum and Petrie. We consider the problem of maximum likelihood estimation of the HMMR parameters and develop analogs for the methods used in HMM's for our regression case. Simulation studies indicate consistency and asymptotic normality of the suggested estimates.