Sequential Monte Carlo smoothing with application to parameter estimation in nonlinear state space models
(2008) In Bernoulli 14(1). p.155-179- Abstract
- This paper concerns the use of sequential Monte Carlo methods (SMC) for smoothing in general state space models. A well-known problem when applying the standard SMC technique in the smoothing mode is that the resampling mechanism introduces degeneracy of the approximation in the path space. However, when performing maximum likelihood estimation via the EM algorithm, all functionals involved are of additive form for a large subclass of models. To cope with the problem in this case, a modification of the standard method (based on a technique proposed by Kitagawa and Sato) is suggested. Our algorithm relies on forgetting properties of the filtering dynamics and the quality of the estimates produced is investigated, both theoretically and via... (More)
- This paper concerns the use of sequential Monte Carlo methods (SMC) for smoothing in general state space models. A well-known problem when applying the standard SMC technique in the smoothing mode is that the resampling mechanism introduces degeneracy of the approximation in the path space. However, when performing maximum likelihood estimation via the EM algorithm, all functionals involved are of additive form for a large subclass of models. To cope with the problem in this case, a modification of the standard method (based on a technique proposed by Kitagawa and Sato) is suggested. Our algorithm relies on forgetting properties of the filtering dynamics and the quality of the estimates produced is investigated, both theoretically and via simulations. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/1271937
- author
- Olsson, Jimmy LU ; Cappé, Olivier ; Douc, Randal and Moulines, Éric
- organization
- publishing date
- 2008
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- stochastic volatility model, exponential family, state space models, particle filters, EM algorithm, sequential Monte Carlo methods
- in
- Bernoulli
- volume
- 14
- issue
- 1
- pages
- 155 - 179
- publisher
- Chapman and Hall
- external identifiers
-
- scopus:41449089271
- ISSN
- 1350-7265
- DOI
- 10.3150/07-BEJ6150
- language
- English
- LU publication?
- yes
- id
- 150c3923-2aaf-477c-be6c-8d4e1caf3147 (old id 1271937)
- date added to LUP
- 2016-04-01 14:38:16
- date last changed
- 2022-03-29 22:03:09
@article{150c3923-2aaf-477c-be6c-8d4e1caf3147,
abstract = {{This paper concerns the use of sequential Monte Carlo methods (SMC) for smoothing in general state space models. A well-known problem when applying the standard SMC technique in the smoothing mode is that the resampling mechanism introduces degeneracy of the approximation in the path space. However, when performing maximum likelihood estimation via the EM algorithm, all functionals involved are of additive form for a large subclass of models. To cope with the problem in this case, a modification of the standard method (based on a technique proposed by Kitagawa and Sato) is suggested. Our algorithm relies on forgetting properties of the filtering dynamics and the quality of the estimates produced is investigated, both theoretically and via simulations.}},
author = {{Olsson, Jimmy and Cappé, Olivier and Douc, Randal and Moulines, Éric}},
issn = {{1350-7265}},
keywords = {{stochastic volatility model; exponential family; state space models; particle filters; EM algorithm; sequential Monte Carlo methods}},
language = {{eng}},
number = {{1}},
pages = {{155--179}},
publisher = {{Chapman and Hall}},
series = {{Bernoulli}},
title = {{Sequential Monte Carlo smoothing with application to parameter estimation in nonlinear state space models}},
url = {{http://dx.doi.org/10.3150/07-BEJ6150}},
doi = {{10.3150/07-BEJ6150}},
volume = {{14}},
year = {{2008}},
}