ONLINE SEQUENTIAL MONTE CARLO EM ALGORITHM

被引：24
作者：
Cappe, Olivier [1 ]
机构：
[1] Telecom ParisTech, LTCI, F-75013 Paris, France
来源：
2009 IEEE/SP 15TH WORKSHOP ON STATISTICAL SIGNAL PROCESSING, VOLS 1 AND 2 | 2009年
关键词：
D O I：
10.1109/SSP.2009.5278646
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
Online (or recursive) estimation of fixed model parameters in general state-space models is a crucial but often difficult task. This paper is about likelihood-based point estimation, showing that an online EM (Expectation-Maximization) algorithm recently proposed for discrete hidden Markov models can be extended to more general settings, including non-linear non-Gaussian state-space models that necessitate the use of sequential Monte Carlo filtering approximations. The performance of the proposed online sequential Monte Carlo EM algorithm is illustrated on numerical examples.
引用
收藏
页码：37 / 40
页数：4
相关论文
共 16 条
[1]  
ANDRIEU C, 2005, P IEEE C DEC CONTR
[2]  
ANDRIEU C, 2003, P IEEE INT C AC SPEE, V6, P69
[3]  
Cappe O., 2009, ONLINE EM ALGORITHM
[4]  
Cappe O., 2005, SPR S STAT
[5]   On-line expectation-maximization algorithm for latent data models [J].
Cappe, Olivier ;
Moulines, Eric .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2009, 71 :593-613
[6]  
Doucet A., 2001, Sequential Monte Carlo methods in practice, V1
[7]  
Elliott R.J., 1995, Hidden Markov Models: Estimation and Control
[8]   Markov chain Monte Carlo, sufficient statistics, and particle filters [J].
Fearnhead, P .
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2002, 11 (04) :848-862
[9]   NOVEL-APPROACH TO NONLINEAR NON-GAUSSIAN BAYESIAN STATE ESTIMATION [J].
GORDON, NJ ;
SALMOND, DJ ;
SMITH, AFM .
IEE PROCEEDINGS-F RADAR AND SIGNAL PROCESSING, 1993, 140 (02) :107-113
[10]  
JOHANSEN A, 2006, P IEEE INT C AC SPEE
12