论文信息 - Prediction in hidden Markov Models using sequential Monte Carlo methods

Prediction in hidden Markov Models using sequential Monte Carlo methods

A novel method of multistep-ahead prediction based on joint probability distribution is proposed in this paper. Firstly, we introduce the basic theory of hidden Markov model (HMM) and sequential Monte Carlo (SMC) method. Secondly, we make the joint multistep-ahead prediction using SMC method in HMM and then develop the corresponding on-line algorithm. At last, the data of monthly national air passengers in America, from Jan. 1990 to Aug 2001, are analyzed, and experimental results demonstrate that the method proposed in this paper is effective.

Xuanxi Ning | Hongwei Ma | Dongqing Zhang | Xueni Liu

[1] Lawrence R. Rabiner,et al. A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[2] Louis A. Liporace,et al. Maximum likelihood estimation for multivariate observations of Markov sources , 1982, IEEE Trans. Inf. Theory.

[3] Neil J. Gordon,et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[4] Simon J. Godsill,et al. On sequential Monte Carlo sampling methods for Bayesian filtering , 2000, Stat. Comput..

[5] Kristine L. Bell,et al. A Tutorial on Particle Filters for Online Nonlinear/NonGaussian Bayesian Tracking , 2007 .

[6] Neil J. Gordon,et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[7] Niclas Bergman,et al. Recursive Bayesian Estimation : Navigation and Tracking Applications , 1999 .

[8] Yuguo Chen,et al. Sequential Monte Carlo methods for filtering and smoothing in hidden Markov models , 2003, IEEE Workshop on Statistical Signal Processing, 2003.