Sequential Monte Carlo smoothing with estimation in non-linear state space models
(2006) In Preprint without journal information- 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 involved functionals will be of additive form for a large subclass of models. To cope with the problem in this case, a modification, relying on forgetting properties of the filtering dynamics, of the standard method is proposed. In this setting, the quality of the produced estimates is investigated both theoretically and through simulations.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/931338
- author
- Olsson, Jimmy LU ; Cappé, Olivier ; Douc, Randal and Moulines, Éric
- organization
- publishing date
- 2006
- type
- Contribution to journal
- publication status
- unpublished
- subject
- in
- Preprint without journal information
- issue
- 2006:15
- publisher
- Manne Siegbahn Institute
- ISSN
- 0348-7911
- language
- English
- LU publication?
- yes
- id
- 8519b3a6-db6c-420e-8931-a538ac05a4f5 (old id 931338)
- date added to LUP
- 2016-04-04 09:40:44
- date last changed
- 2018-11-21 20:54:51
@article{8519b3a6-db6c-420e-8931-a538ac05a4f5, 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 involved functionals will be of additive form for a large subclass of models. To cope with the problem in this case, a modification, relying on forgetting properties of the filtering dynamics, of the standard method is proposed. In this setting, the quality of the produced estimates is investigated both theoretically and through simulations.}}, author = {{Olsson, Jimmy and Cappé, Olivier and Douc, Randal and Moulines, Éric}}, issn = {{0348-7911}}, language = {{eng}}, number = {{2006:15}}, publisher = {{Manne Siegbahn Institute}}, series = {{Preprint without journal information}}, title = {{Sequential Monte Carlo smoothing with estimation in non-linear state space models}}, year = {{2006}}, }