Home > Research > Publications & Outputs > Efficient sequential Monte Carlo with multiple ...

Research

Links

Text available via DOI:

View graph of relations

Efficient sequential Monte Carlo with multiple proposals and control variates

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Efficient sequential Monte Carlo with multiple proposals and control variates. / Li, Wentao; Chen, Rong; Tan, Zhiqiang.
In: Journal of the American Statistical Association, Vol. 111, No. 513, 2016, p. 298-313.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Li, W, Chen, R & Tan, Z 2016, 'Efficient sequential Monte Carlo with multiple proposals and control variates', Journal of the American Statistical Association, vol. 111, no. 513, pp. 298-313. https://doi.org/10.1080/01621459.2015.1006364

APA

Li, W., Chen, R., & Tan, Z. (2016). Efficient sequential Monte Carlo with multiple proposals and control variates. Journal of the American Statistical Association, 111(513), 298-313. https://doi.org/10.1080/01621459.2015.1006364

Vancouver

Li W, Chen R, Tan Z. Efficient sequential Monte Carlo with multiple proposals and control variates. Journal of the American Statistical Association. 2016;111(513):298-313. Epub 2015 Feb 6. doi: 10.1080/01621459.2015.1006364

Author

Li, Wentao ; Chen, Rong ; Tan, Zhiqiang. / Efficient sequential Monte Carlo with multiple proposals and control variates. In: Journal of the American Statistical Association. 2016 ; Vol. 111, No. 513. pp. 298-313.

Bibtex

@article{5108e64cb5b14cb1894012d74aeeec48,
title = "Efficient sequential Monte Carlo with multiple proposals and control variates",
abstract = "Sequential Monte Carlo is a useful simulation-based method for on-line filtering of state space models. For certain complex state space models, a single proposal distribution is usually not satisfactory and using multiple proposal distributions is a general approach to address various aspects of the filtering problem. This paper proposes an efficient method of using multiple proposals in combination with control variates. The likelihood approach of Tan (2004) likelihood is used in both resampling and estimation. The new algorithm is shown to be asymptotically more efficient than the direct use of multiple proposals and control variates. The guidance for selecting multiple proposals and control variates is also given. Numerical examples are used to demonstrate that the new algorithm can significantly improve over the bootstrap filter and auxiliary particle filter.",
author = "Wentao Li and Rong Chen and Zhiqiang Tan",
year = "2016",
doi = "10.1080/01621459.2015.1006364",
language = "English",
volume = "111",
pages = "298--313",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "513",

}

RIS

TY - JOUR

T1 - Efficient sequential Monte Carlo with multiple proposals and control variates

AU - Li, Wentao

AU - Chen, Rong

AU - Tan, Zhiqiang

PY - 2016

Y1 - 2016

N2 - Sequential Monte Carlo is a useful simulation-based method for on-line filtering of state space models. For certain complex state space models, a single proposal distribution is usually not satisfactory and using multiple proposal distributions is a general approach to address various aspects of the filtering problem. This paper proposes an efficient method of using multiple proposals in combination with control variates. The likelihood approach of Tan (2004) likelihood is used in both resampling and estimation. The new algorithm is shown to be asymptotically more efficient than the direct use of multiple proposals and control variates. The guidance for selecting multiple proposals and control variates is also given. Numerical examples are used to demonstrate that the new algorithm can significantly improve over the bootstrap filter and auxiliary particle filter.

AB - Sequential Monte Carlo is a useful simulation-based method for on-line filtering of state space models. For certain complex state space models, a single proposal distribution is usually not satisfactory and using multiple proposal distributions is a general approach to address various aspects of the filtering problem. This paper proposes an efficient method of using multiple proposals in combination with control variates. The likelihood approach of Tan (2004) likelihood is used in both resampling and estimation. The new algorithm is shown to be asymptotically more efficient than the direct use of multiple proposals and control variates. The guidance for selecting multiple proposals and control variates is also given. Numerical examples are used to demonstrate that the new algorithm can significantly improve over the bootstrap filter and auxiliary particle filter.

U2 - 10.1080/01621459.2015.1006364

DO - 10.1080/01621459.2015.1006364

M3 - Journal article

VL - 111

SP - 298

EP - 313

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 513

ER -