Particle Learning and Smoothing

TL;DR

Particle learning offers a unified, efficient approach for state filtering, parameter estimation, and smoothing in state space models, outperforming traditional methods and rivaling MCMC in accuracy.

Contribution

The paper introduces particle learning with a fully-adapted filter using conditional sufficient statistics, enhancing state filtering, parameter learning, and smoothing capabilities.

Findings

PL outperforms existing particle filtering methods

PL is competitive with MCMC in accuracy

State smoothing with parameter uncertainty is effectively addressed

Abstract

Particle learning (PL) provides state filtering, sequential parameter learning and smoothing in a general class of state space models. Our approach extends existing particle methods by incorporating the estimation of static parameters via a fully-adapted filter that utilizes conditional sufficient statistics for parameters and/or states as particles. State smoothing in the presence of parameter uncertainty is also solved as a by-product of PL. In a number of examples, we show that PL outperforms existing particle filtering alternatives and proves to be a competitor to MCMC.