Particle-based likelihood inference in partially observed diffusion processes using generalised Poisson estimators
(2011) In Electronic Journal of Statistics 5. p.1090-1122- Abstract
- This paper concerns the use of the expectation-maximisation (EM) algorithm for inference in partially observed diffusion processes. In this context, a well known problem is that all except a few diffusion processes lack closed-form expressions of the transition densities. Thus, in order to estimate efficiently the EM intermediate quantity we construct, using novel techniques for unbiased estimation of diffusion transition densities, a random weight fixed-lag auxiliary particle smoother, which avoids the well known problem of particle trajectory degeneracy in the smoothing mode. The estimator is justified theoretically and demonstrated on a simulated example.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/2179491
- author
- Olsson, Jimmy LU and Ströjby, Jonas LU
- organization
- publishing date
- 2011
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Auxiliary particle filter, EM algorithm, exact algorithm, generalised, Poisson estimator, partially observed diffusion process, sequential, Monte Carlo
- in
- Electronic Journal of Statistics
- volume
- 5
- pages
- 1090 - 1122
- publisher
- Institute of Mathematical Statistics
- external identifiers
-
- wos:000295324700001
- scopus:84859831900
- ISSN
- 1935-7524
- DOI
- 10.1214/11-EJS632
- language
- English
- LU publication?
- yes
- id
- 5ad4ab8f-f7da-4cc7-b80c-5cd2583e32de (old id 2179491)
- date added to LUP
- 2016-04-01 13:50:09
- date last changed
- 2022-01-27 21:21:41
@article{5ad4ab8f-f7da-4cc7-b80c-5cd2583e32de, abstract = {{This paper concerns the use of the expectation-maximisation (EM) algorithm for inference in partially observed diffusion processes. In this context, a well known problem is that all except a few diffusion processes lack closed-form expressions of the transition densities. Thus, in order to estimate efficiently the EM intermediate quantity we construct, using novel techniques for unbiased estimation of diffusion transition densities, a random weight fixed-lag auxiliary particle smoother, which avoids the well known problem of particle trajectory degeneracy in the smoothing mode. The estimator is justified theoretically and demonstrated on a simulated example.}}, author = {{Olsson, Jimmy and Ströjby, Jonas}}, issn = {{1935-7524}}, keywords = {{Auxiliary particle filter; EM algorithm; exact algorithm; generalised; Poisson estimator; partially observed diffusion process; sequential; Monte Carlo}}, language = {{eng}}, pages = {{1090--1122}}, publisher = {{Institute of Mathematical Statistics}}, series = {{Electronic Journal of Statistics}}, title = {{Particle-based likelihood inference in partially observed diffusion processes using generalised Poisson estimators}}, url = {{http://dx.doi.org/10.1214/11-EJS632}}, doi = {{10.1214/11-EJS632}}, volume = {{5}}, year = {{2011}}, }