Exponential forgetting of smoothing distributions for pairwise Markov models

DOI10.1214/21-EJP628MaRDI QIDQ2042802zbMATH OpenWikidataFDO

Authors J. Lember, Joonas Sova

Publication date 21 July 2021

Published in Electronic Journal of Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2103.05474

zbMATH Keywords

hidden Markov modelsMarkov modelssmoothing probabilities

Mathematics Subject Classification ID

Discrete-time Markov processes on general state spaces (60J05) Computational methods in Markov chains (60J22)

Abstract: We consider a bivariate Markov chain $Z={Z_k}_{kgeq1}={(X_k,Y_k)}_{kgeq1}$ taking values on product space $calZ=calXimescalY$, where $calX$ is possibly uncountable space and $calY=1,ldots,|calY|$ is a finite state-space. The purpose of the paper is to find sufficient conditions that guarantee the exponential convergence of smoothing, filtering and predictive probabilities: ${\displaystyle }$sup_{ngeq t}|P(Y_{t:infty}in cdot|X_{l:n})-P(Y_{t:infty}in cdot|X_{s:n}) |_{ m TV} leq K_s alpha^{t}, quad mbox{a.s.}${\displaystyle }$ Here $tgeqsgeqlgeq1$, $K_s$ is $sigma(X_{s:infty})$-measurable finite random variable and $alphain(0,1)$ is fixed. In the second part of the paper, we establish two-sided versions of the above-mentioned convergence. We show that the desired convergences hold under fairly general conditions. A special case of above-mentioned very general model is popular hidden Markov model (HMM). We prove that in HMM-case, our assumptions are more general than all similar mixing-type of conditions encountered in practice, yet relatively easy to verify.

Recommendations

• scientific article; zbMATH DE number 7424978
• Exponential forgetting and geometric ergodicity in hidden Markov models
• On approximation of smoothing probabilities for hidden Markov models
• Exponential stability of filters and smoothers for hidden Markov models
• Forgetting of the initial condition for the filter in general state-space hidden Markov chain: a coupling approach

Cites work

• A complete solution to Blackwell's unique ergodicity problem for hidden Markov chains
• A Constructive Proof of the Existence of Viterbi Processes
• A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle
• ADJUSTED VITERBI TRAINING
• An improved bound for the exponential stability of predictive filters of hidden Markov models
• Analysis of time series subject to changes in regime
• Asymptotic filtering for finite state Markov chains
• Asymptotic risks of Viterbi segmentation
• Bridging Viterbi and posterior decoding: a generalized risk approach to hidden path inference based on hidden Markov models
• Conditional ergodicity in infinite dimension
• Discrete time nonlinear filters with informative observations are stable
• Entropy and information theory.
• Ergodicity and stability of the conditional distributions of nondegenerate Markov chains
• Ergodicity of Autoregressive Processes with Markov-Switching and Consistency of the Maximum-Likelihood Estimator
• Existence of infinite Viterbi path for pairwise Markov models
• Exponential forgetting and geometric ergodicity in hidden Markov models
• Forgetting of the initial condition for the filter in general state-space hidden Markov chain: a coupling approach
• Forgetting of the initial distribution for nonergodic hidden Markov chains
• Forgetting the initial distribution for hidden Markov models
• scientific article; zbMATH DE number 2144803 (Why is no real title available?)
• scientific article; zbMATH DE number 918233 (Why is no real title available?)
• Inference in hidden Markov models.
• Markov Chains and Stochastic Stability
• Observability and nonlinear filtering
• On approximation of smoothing probabilities for hidden Markov models
• On discrete time ergodic filters with wrong initial data
• On factor maps that send Markov measures to Gibbs measures
• On stability of nonlinear AR processes with Markov switching
• On the accuracy of the MAP inference in HMMs
• On the exchange of intersection and supremum of \({\sigma}\)-fields in filtering theory
• Phase transitions in nonlinear filtering
• Regenerativity of Viterbi process for pairwise Markov models
• Signal and Image Segmentation Using Pairwise Markov Chains
• Stability of nonlinear filters in nonmixing case
• Stability of the nonlinear filter for slowly switching Markov chains
• Stationarity of multivariate Markov-switching ARMA models
• The adjusted Viterbi training for hidden Markov models
• The stability of conditional Markov processes and Markov chains in random environments
• Uniform observability of hidden Markov models and filter stability for unstable signals
• Unsupervised data classification using pairwise Markov chains with automatic copulas selection

Cited in

(3)

• Pairwise Markov models and hybrid segmentation approach
• On approximation of smoothing probabilities for hidden Markov models
• Branch-and-bound method for calculating Viterbi path in triplet Markov models

This page was built for publication: Exponential forgetting of smoothing distributions for pairwise Markov models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2042802)