Approximation of stationary processes by hidden Markov models

DOI10.1007/S00498-010-0050-7MaRDI QIDQ661013zbMATH OpenOpenAlexDBLPWikidataFDO

Authors Lorenzo Finesso, Angela Grassi, Peter Spreij

Publication date 6 February 2012

Published in MCSS. Mathematics of Control, Signals, and Systems (Search for Journal in Brave)

Copyright license Creative Commons Attribution-NonCommercial 4.0 International

Full work available at URL https://arxiv.org/abs/math/0606591

zbMATH Keywords

approximationdivergence rateHankel matrixhidden Markov model (HMM)Kullback-Leibler divergencenonnegative matrix factorization (NMF)

Mathematics Subject Classification ID

Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Stochastic systems in control theory (general) (93E03)

Abstract: We aim at the construction of a Hidden Markov Model (HMM) of assigned complexity (number of states of the underlying Markov chain) which best approximates, in Kullback-Leibler divergence rate, a given stationary process. We establish, under mild conditions, the existence of the divergence rate between a stationary process and an HMM. Since in general there is no analytic expression available for this divergence rate, we approximate it with a properly defined, and easily computable, divergence between Hankel matrices, which we use as our approximation criterion. We propose a three-step algorithm, based on the Nonnegative Matrix Factorization technique, which realizes an HMM optimal with respect to the defined approximation criterion. A full theoretical analysis of the algorithm is given in the special case of Markov approximation.

Recommendations

• Stationary distribution estimation in hidden Markov models
• scientific article; zbMATH DE number 2144803
• On approximation of smoothing probabilities for hidden Markov models
• Approximating a diffusion by a finite-state hidden Markov model
• Almost sure parameter estimation and convergence rates for hidden Markov models
• Normal approximation for functions of hidden Markov models
• Hidden Markov processes
• Limit Theorems in Hidden Markov Models
• Parameter estimation for continuous time hidden Markov processes
• Maximum likelihood estimation of hidden Markov processes

Cites work

• Analyticity of Entropy Rate of Hidden Markov Chains
• Asymptotical Statistics of Misspecified Hidden Markov Models
• Exponential forgetting and geometric ergodicity in hidden Markov models
• scientific article; zbMATH DE number 3872359 (Why is no real title available?)
• scientific article; zbMATH DE number 48436 (Why is no real title available?)
• I-divergence geometry of probability distributions and minimization problems
• Learning the parts of objects by non-negative matrix factorization
• Markov Chains
• Maximum-likelihood estimation for hidden Markov models
• Nonnegative matrix factorization and I-divergence alternating minimization
• On Stochastic Processes Derived From Markov Chains
• On the convergence properties of the EM algorithm
• Statistical Inference for Probabilistic Functions of Finite State Markov Chains
• The realization problem for hidden Markov models

Cited in

(7)

• Approximating a diffusion by a finite-state hidden Markov model
• Equivalence of state representations for hidden Markov models
• Estimation of hidden frequencies for 2D stationary process
• Optimal Kullback-Leibler approximation of Markov chains via nuclear norm regularisation
• A correction to ``On approximation of smoothing probabilities for hidden Markov models
• Limits of learning dynamical systems
• Structured nonnegative matrix factorization with applications to hidden Markov realization and clustering

This page was built for publication: Approximation of stationary processes by hidden Markov models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q661013)