Online expectation-maximization type algorithms for parameter estimation in general state space models

C Andrieu,A Doucet

doi:10.1109/icassp.2003.1201620

Online expectation-maximization type algorithms for parameter estimation in general state space models

C Andrieu, A Doucet

https://doi.org/10.1109/icassp.2003.1201620

Copy DOI

Publication Date: Apr 6, 2003

Citations: 58

Affiliation: University of Bristol, University of Cambridge

#Nonlinear non-Gaussian State Space Models #Standard Sequential Monte Carlo + Show 8 more

Abstract
Full-Text PDF
Similar Papers

Abstract

We present new online algorithms to estimate static parameters in nonlinear non-Gaussian state space models. These algorithms rely on online expectation-maximization (EM) type algorithms. Contrary to standard sequential Monte Carlo (SMC) methods recently proposed in the literature, these algorithms do not degenerate over time.

Full Text