Time discretization of continuous-time filters and smoothers for HMM parameter estimation

@article{James1996TimeDO,
  title={Time discretization of continuous-time filters and smoothers for HMM parameter estimation},
  author={Matthew R. James and Vikram Krishnamurthy and François Le Gland},
  journal={IEEE Trans. Inf. Theory},
  year={1996},
  volume={42},
  pages={593-605}
}
In this paper we propose algorithms for parameter estimation of fast-sampled homogeneous Markov chains observed in white Gaussian noise. Our algorithms are obtained by the robust discretization of stochastic differential equations involved in the estimation of continuous-time hidden Markov models (HMM's) via the EM algorithm. We present two algorithms: the first is based on the robust discretization of continuous-time filters that were recently obtained by Elliott to estimate quantities used in… 

Figures from this paper

Data-Recursive Smoother Formulae for Partially Observed Discrete-Time Markov Chains

TLDR
This article considers HMM parameter estimation in the context of a filter and smoother based expectation maximization (EM) algorithms, which require the entire log likelihood function to be recalculated upon the arrival of new measurements.

Moment Based Regression Algorithms for Drift and Volatility Estimation in Continuous-Time Markov Switching Models

TLDR
The moments of the observed process are computed explicitly as a function of the time discretization interval of the discrete time observation process and a method of moments based parameter estimator is proposed.

EM algorithm for Markov chains observed via Gaussian noise and point process information: Theory and case studies

Abstract In this paper we study parameter estimation via the Expectation Maximization (EM) algorithm for a continuous-time hidden Markov model with diffusion and point process observation. Inference

General smoothing formulas for Markov-modulated Poisson observations

TLDR
This paper considers two model classes, each Markov modulated Poisson processes, whose stochastic intensities depend upon the state of an unobserved Markov process, and compute general smoothing dynamics for partially observed dynamical systems generating Poisson observations.

Filtering, Smoothing and M-ary Detection with Discrete Time Poisson Observations

Abstract In this article, we solve a class of estimation problems, namely, filtering smoothing and detection for a discrete time dynamical system with integer-valued observations. The observation

Robust M-ary detection filters and smoothers for continuous-time jump Markov systems

TLDR
A dynamic M-ary detection problem when Markov chains are observed through a Wiener process is considered and robust matrix-valued dynamics, whose solutions estimate joint probabilities for all combinations of candidate model parameter sets and values taken by the indirectly observed state process are computed.

Discrete-Time Expectation Maximization Algorithms for Markov-Modulated Poisson Processes

TLDR
Using the expectation maximization algorithm framework, filters and smoothers can be applied to estimate the parameters of the model in either an online configuration or an offline configuration and the estimator dynamics do not involve stochastic integrals.

ℒ1-Optimal Filtering of Markov Jump Processes. I. Exact Solution and Numerical Implementation Schemes

  • A. Borisov
  • Mathematics, Computer Science
    Autom. Remote. Control.
  • 2020
TLDR
The main results are the statements on the accuracy of the approximate solution of the filtering problem, depending on the number of jumps taken into account for the estimated state, on the discretization step, and on the numerical integration scheme applied.

Robust continuous-time smoothers-without two-sided stochastic integrals

We consider the problem of fixed-interval smoothing of a continuous-time partially observed nonlinear stochastic dynamical system. Existing results for such smoothers require the use of two sided

Studies in nonlinear filtering theory: random parameter linear systems, target tracking and communication constrained estimation

TLDR
The focus of this thesis is nonlinear ltering for discrete-time stochastic systems and state estimation for Markov chains in the presence of communication constraints and optimal and suboptimal ltering algorithms for random parameter linear systems.
...

References

SHOWING 1-10 OF 22 REFERENCES

Exact adaptive filters for Markov chains observed in Gaussian noise

Exact filters for the estimation of the number of transitions of finite-state continuous-time Markov processes

TLDR
A finite-dimensional exact filter is derived, and using the EM algorithm, an application is made to the problem of estimating the unknown transition matrix of a finite-state continuous-time Markov process.

New finite-dimensional filters and smoothers for noisily observed Markov chains

TLDR
Finite-dimensional, recursive filters and smoothers are given for the number of jumps from state i to state j, for the occupation time of state i, and for a stochastic integral related to the drift in the observations.

An introduction to the application of the theory of probabilistic functions of a Markov process to automatic speech recognition

TLDR
This paper presents several of the salient theoretical and practical issues associated with modeling a speech signal as a probabilistic function of a (hidden) Markov chain, and focuses on a particular class of Markov models, which are especially appropriate for isolated word recognition.

Statistics of random processes

1. Essentials of Probability Theory and Mathematical Statistics.- 2. Martingales and Related Processes: Discrete Time.- 3. Martingales and Related Processes: Continuous Time.- 4. The Wiener Process,

Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper

Vibratory power unit for vibrating conveyers and screens comprising an asynchronous polyphase motor, at least one pair of associated unbalanced masses disposed on the shaft of said motor, with the

Some applications of stochastic difierential equations to optimal nonlinear ltering

Stochastic differential equations specifying dynamical structure of filters generating posterior probability distribution when inputs are time functions

A tutorial on hidden Markov models and selected applications in speech recognition

The fabric comprises a novel type of netting which will have particular utility in screening out mosquitoes and like insects and pests. The fabric is defined of voids having depth as well as width

Equations du lissage non lineaire

Le but de cette note est d'etablir trois couples d'equations, chacun des trois permettant de caracteriser la loi conditionnelle d'un probleme de lissage non lineaire.