Approximate methods for state-space models

DOI10.1198/JASA.2009.TM08326MaRDI QIDQ5254945zbMATH OpenOpenAlexWikidataFDO

Authors Shinsuke Koyama, Lucia Castellanos Pérez-Bolde, Cosma Rohilla Shalizi, Robert E. Kass

Publication date 11 June 2015

Published in Journal of the American Statistical Association (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1004.3476, http://europepmc.org/articles/pmc3132892

zbMATH Keywords

Laplace's methodneural decodingrecursive Bayesian estimation

Mathematics Subject Classification ID

Signal detection and filtering (aspects of stochastic processes) (60G35) Bayesian inference (62F15)

Abstract: State-space models provide an important body of techniques for analyzing time-series, but their use requires estimating unobserved states. The optimal estimate of the state is its conditional expectation given the observation histories, and computing this expectation is hard when there are nonlinearities. Existing filtering methods, including sequential Monte Carlo, tend to be either inaccurate or slow. In this paper, we study a nonlinear filter for nonlinear/non-Gaussian state-space models, which uses Laplace's method, an asymptotic series expansion, to approximate the state's conditional mean and variance, together with a Gaussian conditional distribution. This {em Laplace-Gaussian filter} (LGF) gives fast, recursive, deterministic state estimates, with an error which is set by the stochastic characteristics of the model and is, we show, stable over time. We illustrate the estimation ability of the LGF by applying it to the problem of neural decoding and compare it to sequential Monte Carlo both in simulations and with real data. We find that the LGF can deliver superior results in a small fraction of the computing time.

Recommendations

• Monte Carlo Estimation for Nonlinear Non-Gaussian State Space Models
• scientific article; zbMATH DE number 1222386
• Constructing Non-linear Gaussian Time Series by Means of a Simplified State Space Representation
• Monte Carlo Smoothing for Nonlinear Time Series
• State space and hidden Markov models

Cited in

(24)

• Recasting brain-machine interface design from a physical control system perspective
• Deep state-space Gaussian processes
• Markov-switching state space models for uncovering musical interpretation
• Efficient state-space inference of periodic latent force models
• scientific article; zbMATH DE number 4195290 (Why is no real title available?)
• Decoding of neural data using cohomological feature extraction
• Fast estimation methods for time-series models in state–space form
• scientific article; zbMATH DE number 4091302 (Why is no real title available?)
• Approximate Inference in State-Space Models With Heavy-Tailed Noise
• Marginalized approximate filtering of state‐space models
• Statistical algorithms for models in state space using SsfPack 2.2
• Review of SsfPack 2.2: statistical algorithms for models in state space
• The Discriminative Kalman Filter for Bayesian Filtering with Nonlinear and Nongaussian Observation Models
• Comparison of brain-computer interface decoding algorithms in open-loop and closed-loop control
• A new look at state-space models for neural data
• Information transmission using non-Poisson regular firing
• Variational latent Gaussian process for recovering single-trial dynamics from population spike trains
• STABLE ALGORITHMS FOR THE STATE SPACE MODEL
• Semi-Complete Data Augmentation for Efficient State Space Model Fitting
• A point mass proposal method for Bayesian state-space model fitting
• Bellman filtering and smoothing for state-space models
• A probabilistic diagnostic for Laplace approximations: introduction and experimentation
• Title not available (Why is no real title available?)
• Bayesian decoding of neural spike trains

This page was built for publication: Approximate methods for state-space models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5254945)