Fast and numerically stable particle-based online additive smoothing: the AdaSmooth algorithm
Authors:
- Alessandro MastrototaroAMAlessandro Mastrototaro
- Jimmy OlssonJOJimmy Olsson
- Johan AlenlövJAJohan Alenlöv
Preprint
Abstract
We present a novel sequential Monte Carlo approach to online smoothing of additive functionals in a very general class of path-space models. Hitherto, the solutions proposed in the literature suffer from either long-term numerical instability due to particle-path degeneracy or, in the case that degeneracy is remedied by particle approximation of the so-called backward kernel, high computational demands. In order to balance optimally computational speed against numerical stability, we propose to furnish a (fast) naive particle smoother, propagating recursively a sample of particles and associated smoothing statistics, with an adaptive backward-sampling-based updating rule which allows the number of (costly) backward samples to be kept at a minimum. This yields a new, function-specific additive smoothing algorithm, AdaSmooth, which is computationally fast, numerically stable and easy to implement. The algorithm is provided with rigorous theoretical results guaranteeing its consistency, asymptotic normality and long-term stability as well as numerical results demonstrating empirically the clear superiority of AdaSmooth to existing algorithms.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Asymptotic genealogies of interacting particle systems with an application to sequential Monte CarloThe Annals of Statistics, 2020
- On Particle Methods for Parameter Estimation in State-Space ModelsStatistical Science, 2015
- Sequential Monte Carlo smoothing for general state space hidden Markov modelsThe Annals of Applied Probability, 2011
- Online EM Algorithm for Hidden Markov ModelsJournal of Computational and Graphical Statistics, 2011
- A backward particle interpretation of Feynman-Kac formulaeESAIM: Mathematical Modelling and Numerical Analysis, 2010
- Limit theorems for weighted samples with applications to sequential Monte Carlo methodsThe Annals of Statistics, 2008
- Online Learning with Hidden Markov ModelsNeural Computation, 2008
- Monte Carlo Smoothing for Nonlinear Time SeriesJournal of the American Statistical Association, 2004
- On sequential Monte Carlo sampling methods for Bayesian filteringStatistics and Computing, 2000
- Metropolized independent sampling with comparisons to rejection sampling and importance samplingStatistics and Computing, 1996