Document Actions

BWPEF 1607
Demian Pouzo, Zacharias Psaradakis and Martin Sola
Maximum Likelihood Estimation in Possibly Misspecified Dynamic Models with Time-Inhomogeneous Markov Regimes

Abstract
This paper considers maximum likelihood (ML) estimation in a large class of models with hidden Markov regimes. We investigate consistency and local asymptotic normality of the ML estimator under general conditions which allow for autoregressive dynamics in the observable process, time-inhomogeneous Markov regime sequences, and possible model misspecification. A Monte Carlo study examines the finite-sample properties of the ML estimator. An empirical application is also discussed.

Keywords: Autoregressive model, consistency, hidden Markov model, Markov regimes, maximum likelihood, local asymptotic normality, misspecified models, time-inhomogenous Markov chain.

View this paper.