Home > Research > Publications & Outputs > On-Line Inference for Hidden Markov Models via ...

Research

Links

Text available via DOI:

View graph of relations

On-Line Inference for Hidden Markov Models via Particle Filters.

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

Published

Standard

On-Line Inference for Hidden Markov Models via Particle Filters. / Fearnhead, Paul; Clifford, Peter.
In: Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 65, No. 4, 11.2003, p. 887-899.

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

Harvard

Fearnhead, P & Clifford, P 2003, 'On-Line Inference for Hidden Markov Models via Particle Filters.', Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 65, no. 4, pp. 887-899. https://doi.org/10.1111/1467-9868.00421

APA

Fearnhead, P., & Clifford, P. (2003). On-Line Inference for Hidden Markov Models via Particle Filters. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 65(4), 887-899. https://doi.org/10.1111/1467-9868.00421

Vancouver

Fearnhead P, Clifford P. On-Line Inference for Hidden Markov Models via Particle Filters. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2003 Nov;65(4):887-899. doi: 10.1111/1467-9868.00421

Author

Fearnhead, Paul; ; Clifford, Peter. / On-Line Inference for Hidden Markov Models via Particle Filters. In: Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2003 ; Vol. 65, No. 4. pp. 887-899.

Bibtex

@article{01284599578e4f96b97de5d19cc6098b,
title = "On-Line Inference for Hidden Markov Models via Particle Filters.",
abstract = "We consider the on-line Bayesian analysis of data by using a hidden Markov model, where inference is tractable conditional on the history of the state of the hidden component. A new particle filter algorithm is introduced and shown to produce promising results when analysing data of this type. The algorithm is similar to the mixture Kalman filter but uses a different resampling algorithm. We prove that this resampling algorithm is computationally efficient and optimal, among unbiased resampling algorithms, in terms of minimizing a squared error loss function. In a practical example, that of estimating break points from well-log data, our new particle filter outperforms two other particle filters, one of which is the mixture Kalman filter, by between one and two orders of magnitude.",
author = "Paul; Fearnhead and Peter Clifford",
year = "2003",
month = nov,
doi = "10.1111/1467-9868.00421",
language = "English",
volume = "65",
pages = "887--899",
journal = "Journal of the Royal Statistical Society: Series B (Statistical Methodology)",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "4",

}

RIS

TY - JOUR

T1 - On-Line Inference for Hidden Markov Models via Particle Filters.

AU - Fearnhead, Paul;

AU - Clifford, Peter

PY - 2003/11

Y1 - 2003/11

N2 - We consider the on-line Bayesian analysis of data by using a hidden Markov model, where inference is tractable conditional on the history of the state of the hidden component. A new particle filter algorithm is introduced and shown to produce promising results when analysing data of this type. The algorithm is similar to the mixture Kalman filter but uses a different resampling algorithm. We prove that this resampling algorithm is computationally efficient and optimal, among unbiased resampling algorithms, in terms of minimizing a squared error loss function. In a practical example, that of estimating break points from well-log data, our new particle filter outperforms two other particle filters, one of which is the mixture Kalman filter, by between one and two orders of magnitude.

AB - We consider the on-line Bayesian analysis of data by using a hidden Markov model, where inference is tractable conditional on the history of the state of the hidden component. A new particle filter algorithm is introduced and shown to produce promising results when analysing data of this type. The algorithm is similar to the mixture Kalman filter but uses a different resampling algorithm. We prove that this resampling algorithm is computationally efficient and optimal, among unbiased resampling algorithms, in terms of minimizing a squared error loss function. In a practical example, that of estimating break points from well-log data, our new particle filter outperforms two other particle filters, one of which is the mixture Kalman filter, by between one and two orders of magnitude.

U2 - 10.1111/1467-9868.00421

DO - 10.1111/1467-9868.00421

M3 - Journal article

VL - 65

SP - 887

EP - 899

JO - Journal of the Royal Statistical Society: Series B (Statistical Methodology)

JF - Journal of the Royal Statistical Society: Series B (Statistical Methodology)

SN - 1369-7412

IS - 4

ER -