Monte Carlo approximations for general state-space models

被引:97
作者
Hurzeler, M [1 ]
Kunsch, HR [1 ]
机构
[1] ETH Zentrum, Seminar Stat, CH-8092 Zurich, Switzerland
关键词
Kalman filter; Kalman smoother; Monte Carlo methods; nonlinear time series analysis; robustness for time series; state-space models;
D O I
10.2307/1390812
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Nonlinear and non-Gaussian state-space models form a large and flexible model class in time series analysis. Two methods for sequentially generating samples from filter densities and smoother densities by simple rejection algorithms are introduced. We illustrate the behavior of our methods in several nonlinear and non-Gaussian examples and compare them with other well-known methods.
引用
收藏
页码:175 / 193
页数:19
相关论文
共 15 条
[1]   A MONTE-CARLO APPROACH TO NONNORMAL AND NONLINEAR STATE-SPACE MODELING [J].
CARLIN, BP ;
POLSON, NG ;
STOFFER, DS .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1992, 87 (418) :493-500
[2]  
Fruhwirth-Schnatter S., 1994, Journal of Time Series Analysis, V15, P183, DOI [10.1111/j.1467-9892.1994.tb00184.x, DOI 10.1111/J.1467-9892.1994.TB00184.X]
[3]   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
[4]  
HARRISON PJ, 1976, J R STAT SOC B, V38, P205
[5]  
Huber P.J., 1981, ROBUST STAT
[7]  
Kitagawa G., 1996, Journal of Computational and Graphical Statistics, V5, P1, DOI [10.1080/10618600.1996.10474692, DOI 10.1080/10618600.1996.10474692, DOI 10.2307/1390750]
[8]  
MARTIN R. D, 1980, Applied time series analysis II, VII, P683
[9]  
NETTO MLA, 1978, IEEE T AUTOMATIC CON, V23, P1062
[10]  
Ripley B.D., 1987, Stochastic Simulation