×

Parameter estimation and asymptotic stability in stochastic filtering. (English) Zbl 1097.62093

Summary: We study the problem of estimating a Markov chain \(X\) (signal) from its noisy partial information Y, when the transition probability kernel depends on some unknown parameters. Our goal is to compute the conditional distribution process \(\mathbb {P}\{X_{n}\,|\, Y_{n}, \dots, Y_{n}\}\), referred to hereafter as the optimal filter. Following a standard Bayesian technique, we treat the parameters as a non-dynamic component of the Markov chain. As a result, the new Markov chain is not going to be mixing, even if the original one is. We show that, under certain conditions, the optimal filters are still going to be asymptotically stable with respect to the initial conditions. Thus, by computing the optimal filter of the new system, we can estimate the signal adaptively.

MSC:

62M20 Inference from stochastic processes and prediction
62M05 Markov processes: estimation; hidden Markov models
62F15 Bayesian inference
60G35 Signal detection and filtering (aspects of stochastic processes)
93D20 Asymptotic stability in control theory
62F12 Asymptotic properties of parametric estimators
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Andrieu, C.; Doucet, A., Particle filtering for partially observed Gaussian state space models, J. R. Stat. Soc. B (methodological), 64, 4, 827-836 (2002) · Zbl 1067.62098
[2] Atar, R.; Zeitouni, O., Exponential stability for nonlinear filtering, Ann. Inst. H. Poincaré Probab. Statist., 33, 6, 697-725 (1997) · Zbl 0888.93057
[3] Baxendale, P.; Chigansky, P.; Lipster, R., Asymptotic stability of the Wonham filter: ergodic and nonergodic signals, SIAM J. Control Optim., 43, 2, 643-669 (2004) · Zbl 1101.93074
[4] Budhiraja, A., Asymptotic stability, ergodicity and other asymptotic properties of the nonlinear filter, Ann. Inst. H. Poincaré Probab. Statist., 39, 6, 919-941 (2003) · Zbl 1160.60318
[5] Budhiraja, A.; Ocone, D., Exponential Stability in discrete-time filtering for non-ergodic signals, Stoch. Proc. Appl., 82, 245-257 (1999) · Zbl 1056.93065
[6] Cérou, F., Long time behavior for some dynamical noise free nonlinear problems, SIAM J. Control Optim., 38, 4, 1086-1101 (2000) · Zbl 0979.93116
[7] Cheney, W., Analysis for Applied Mathematics (2001), Springer-Verlag: Springer-Verlag New York · Zbl 0984.46006
[8] Chigansky, P.; Lipster, R., Stability of nonlinear filters in non-mixing case, Ann. Appl. Probab., 14, 4, 2038-2056 (2004) · Zbl 1065.93034
[9] Crisan, D.; Del Moral, P.; Lyons, T., Discrete filtering using branching and interacting particle systems, Markov Process. Related Fields, 5, 3, 293-318 (1999) · Zbl 0967.93088
[10] Crisan, D.; Doucet, A., A survey of convergence results on particle filtering methods for practitioners, IEEE Trans. Signal Process., 50, 3, 736-746 (2002) · Zbl 1369.60015
[11] Del Moral, P., A uniform convergence theorem for numerical solving of the nonlinear filtering problem, J. Appl. Probab., 35, 4, 873-884 (1998) · Zbl 0940.60060
[12] Del Moral, P.; Guionnet, A., On the stability of measure valued processes with applications to filtering, C. R. Acad. Sci. Paris. Série I. Mathématique, 329, 429-434 (1999) · Zbl 0935.92001
[13] Del Moral, P.; Miclo, L., Branching and interacting particle systems approximations of Feynman-Kac formulae with application to non-linear filtering, (Azéma, J.; Émery, M.; Ledoux, M.; Yor, M., Séminaire de Probabilités XXXIV. Séminaire de Probabilités XXXIV, Lecture Notes in Mathematics, vol. 1729 (2000), Springer-Verlag: Springer-Verlag Berlin), 1-145 · Zbl 0963.60040
[14] Douc, R.; Moulines, E.; Ryden, T., Asymptotic properties of the maximum likelihood estimator in autoregressive models with Markov regime, Ann. Statist., 32, 5, 2254-2304 (2004) · Zbl 1056.62028
[15] Dupuis, P.; Ellis, R. S., A Weak Convergence Approach to the Theory of Large Deviations (1997), John Wiley & Sons, Inc.: John Wiley & Sons, Inc. New York · Zbl 0904.60001
[16] Glötzl, E.; Wakolbinger, A., Bayes estimators and ergodic decomposability with an application to Cox processes, Ann. Probab., 10, 3, 872-876 (1980) · Zbl 0489.62009
[17] Gordon, N. J.; Salmond, D. J.; Smith, A. F.M., Novel approach to nonlinear/non Gaussian Bayesian state estimation, IEE Proc. F, 140, 107-113 (1993)
[18] Kitagawa, G., Monte-Carlo filter and smoother for non-Gaussian nonlinear state space models, J. Comput. Graph. Statist., 5, 1, 1-25 (1996)
[19] Kunita, H., Asymptotic behavior of the nonlinear filtering errors of Markov processes, J. Multivariate Anal., 1, 365-393 (1971)
[20] Le Gland, F.; Oudjane, N., Stability and uniform approximation of nonlinear filters using the Hilbert metric, and application to particle filters, Ann. Appl. Probab., 14, 1, 144-187 (2004) · Zbl 1060.93094
[21] Liu, J.; West, M., Combined parameter and state estimation in simulation-based filtering, (Doucet, A.; de Freitas, N.; Gordon, N., Sequential Monte Carlo in Practice (2001), Springer-Verlag), 197-223 · Zbl 1056.93583
[22] Ocone, D.; Pardoux, E., Asymptotic stability of the optimal filter with respect to its initial condition, SIAM J. Control Optim., 34, 226-243 (1996) · Zbl 1035.93508
[23] Papavasiliou, A., A uniformly convergent adaptive particle filter, J. Appl. Probab., 42, 4, 1053-1068 (2005) · Zbl 1102.60037
[24] Rydén, T., On recursive estimation for hidden Markov models, Stoch. Proc. Appl., 66, 79-96 (1997) · Zbl 0885.62096
[25] L. Stettner, On invariant measures of filtering processes, in: K. Helmes, N. Kohlmann (Eds.), Stochastic Differential Systems, Proc. 4th Bad Honnef Conf., in: Lecture Notes in Control and Inform. Sci., 1989, pp. 279-292; L. Stettner, On invariant measures of filtering processes, in: K. Helmes, N. Kohlmann (Eds.), Stochastic Differential Systems, Proc. 4th Bad Honnef Conf., in: Lecture Notes in Control and Inform. Sci., 1989, pp. 279-292 · Zbl 0683.93082
[26] Storvik, G., Particle filters for state space models with the presence of unknown static parameters, IEEE Trans. Signal Process., 50, 2, 281-289 (2002)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.