Ergodicity of hidden Markov models

In this paper we study ergodic properties of hidden Markov models with a generalized observation structure. In particular sufficient conditions for the exis- tence of a unique invariant measure for the pair filter-observation are given. Fur- thermore, necessary and sufficient conditions for the existence of a unique invariant measure of the triple state-observation-filter are provided in terms of asymptotic stability in probability of incorrectly initialized filters. We also study the asymp- totic properties of the filter and of the state estimator based on the observations as well as on the knowledge of the initial state. Their connection with minimal and maximal invariant measures is also studied.

[1]  H. Kunita.  Ergodic Properties of Nonlinear Filtering Processes , 1991 .

[2]  L. Stettner.  On invariant measures of filtering processes , 1989 .

[3]  Richard L. Tweedie,et al.  Markov Chains and Stochastic Stability , 1993, Communications and Control Engineering Series.

[4]  R. Atar,et al.  Lyapunov Exponents for Finite State Nonlinear Filtering , 1997 .

[5]  Laurent Mevel,et al.  Exponential Forgetting and Geometric Ergodicity in Hidden Markov Models , 2000, Math. Control. Signals Syst..

[6]  Pavel Chigansky,et al.  Asymptotic Stability of the Wonham Filter: Ergodic and Nonergodic Signals , 2002, SIAM J. Control. Optim..

[7]  R. Khasminskii.  Stochastic Stability of Differential Equations , 1980 .

[8]  L. Stettner,et al.  Approximations of discrete time partially observed control problems , 1994 .

[9]  D. Ocone,et al.  Asymptotic Stability of the Optimal Filter with Respect toIts Initial Condition , 1996 .

[10]  Thomas Kaijser.  A Limit Theorem for Partially Observed Markov Chains , 1975 .

[11]  Daniel Hernández-Hernández,et al.  Risk sensitive control of finite state Markov chains in discrete time, with applications to portfolio management , 1999, Math. Methods Oper. Res..

[12]  R. Douc,et al.  Asymptotics of the maximum likelihood estimator for general hidden Markov models , 2001 .

[13]  H. Kunita.  Asymptotic behavior of the nonlinear filtering errors of Markov processes , 1971 .

[14]  Neri Merhav,et al.  Hidden Markov processes , 2002, IEEE Trans. Inf. Theory.

[15]  S. Pliska,et al.  Risk-Sensitive Dynamic Asset Management , 1999 .

[16]  Amarjit Budhiraja,et al.  Asymptotic stability, ergodicity and other asymptotic properties of the nonlinear filter , 2003 .

[17]  L. Rogers.  Stochastic differential equations and diffusion processes: Nobuyuki Ikeda and Shinzo Watanabe North-Holland, Amsterdam, 1981, xiv + 464 pages, Dfl.175.00 , 1982 .

[18]  A. Budhiraja.  Ergodic properties of the nonlinear filter , 2001 .

[19]  Harold J. Kushner,et al.  Monte Carlo Algorithms and Asymptotic Problems in Nonlinear Filtering , 2001 .

[20]  Vivek S. Borkar,et al.  Ergodic control of partially observed Markov chains , 1998 .