Saved in:
Bibliographic Details
Main Authors: Xue, Liwen, Finke, Axel, Johansen, Adam M.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.00696
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916875776032768
author Xue, Liwen
Finke, Axel
Johansen, Adam M.
author_facet Xue, Liwen
Finke, Axel
Johansen, Adam M.
contents We introduce methodology for real-time inference in general-state-space hidden Markov models. Specifically, we extend recent advances in controlled sequential Monte Carlo (CSMC) methods-originally proposed for offline smoothing-to the online setting via a rolling window mechanism. Our novel online rolling controlled sequential Monte Carlo (ORCSMC) algorithm employs two particle systems to simultaneously estimate twisting functions and perform filtering, ensuring real-time adaptivity to new observations while maintaining bounded computational cost. Numerical results on linear-Gaussian, stochastic volatility, and neuroscience models demonstrate improved estimation accuracy and robustness in higher dimensions, compared to standard particle filtering approaches. The method offers a statistically efficient and practical solution for sequential and real-time inference in complex latent variable models.
format Preprint
id arxiv_https___arxiv_org_abs_2508_00696
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Online Rolling Controlled Sequential Monte Carlo
Xue, Liwen
Finke, Axel
Johansen, Adam M.
Computation
We introduce methodology for real-time inference in general-state-space hidden Markov models. Specifically, we extend recent advances in controlled sequential Monte Carlo (CSMC) methods-originally proposed for offline smoothing-to the online setting via a rolling window mechanism. Our novel online rolling controlled sequential Monte Carlo (ORCSMC) algorithm employs two particle systems to simultaneously estimate twisting functions and perform filtering, ensuring real-time adaptivity to new observations while maintaining bounded computational cost. Numerical results on linear-Gaussian, stochastic volatility, and neuroscience models demonstrate improved estimation accuracy and robustness in higher dimensions, compared to standard particle filtering approaches. The method offers a statistically efficient and practical solution for sequential and real-time inference in complex latent variable models.
title Online Rolling Controlled Sequential Monte Carlo
topic Computation
url https://arxiv.org/abs/2508.00696