Likelihood computation for hidden Markov models via generalized two-filter smoothing

DOI10.1016/j.spl.2013.02.005zbMath1417.62234OpenAlexW2963238530MaRDI QIDQ385121

Adam Persing, Ajay Jasra

Publication date: 29 November 2013

Published in: Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.spl.2013.02.005

zbMATH Keywords

sequential Monte Carlolikelihood estimationgeneralized two-filter smoothing

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Inference from stochastic processes and prediction (62M20) Central limit and other weak theorems (60F05) Markov processes: estimation; hidden Markov models (62M05)

Related Items

Exact and approximate Bayesian inference for low integer-valued time series models with intractable likelihoods ⋮ Theory of segmented particle filters ⋮ On the two-filter approximations of marginal smoothing distributions in general state-space models

Cites Work

• Unnamed Item
• Linear variance bounds for particle approximations of time-homogeneous Feynman-Kac formulae
• Sequential Monte Carlo smoothing for general state space hidden Markov models
• A nonasymptotic theorem for unnormalized Feynman-Kac particle models
• Smoothing algorithms for state-space models
• Inference in hidden Markov models.
• A sequential smoothing algorithm with linear computational cost
• Two-filter formulae for discrete-time non-linear bayesian smoothing
• Particle Markov Chain Monte Carlo Methods
• A backward particle interpretation of Feynman-Kac formulae