Approximation of nonnegative systems by moving averages of fixed order

DOI10.1016/J.AUTOMATICA.2019.05.007zbMATH Open1429.93060arXiv1810.10135OpenAlexW2962697573WikidataQ127806644 ScholiaQ127806644MaRDI QIDQ2280796FDOQ2280796

Authors: Lorenzo Finesso, Peter Spreij 

Publication date: 19 December 2019

Published in: Automatica (Search for Journal in Brave)

Abstract: We pose the approximation problem for scalar nonnegative input-output systems via impulse response convolutions of finite order, i.e. finite order moving averages, based on repeated observations of input/output signal pairs. The problem is converted into a nonnegative matrix factorization with special structure for which we use Csisz'ar's I-divergence as the criterion of optimality. Conditions are given, on the input/output data, that guarantee the existence and uniqueness of the minimum. We propose an algorithm of the alternating minimization type for I-divergence minimization, and present its asymptotic behavior. For the case of noisy observations we give the large sample properties of the statistical version of the minimization problem for different observation regimes. Numerical experiments confirm the asymptotic results and exhibit fast convergence of the proposed algorithm.

Full work available at URL: https://arxiv.org/abs/1810.10135

Recommendations

• Optimal nonnegative definite approximations of estimated moving average covariance sequences
• An approximate maximum likelihood estimation for non-Gaussian non-minimum phase moving average processes
• Nonnegative matrix factorization and I-divergence alternating minimization
• On the best quadratic approximation of nonlinear systems
• Structured nonnegative matrix factorization with applications to hidden Markov realization and clustering

zbMATH Keywords

consistencyestimationpositive systemmoving averageinformational divergence

Mathematics Subject Classification ID

Discrete-time control/observation systems (93C55) Realizations from input-output data (93B15) Positive control/observation systems (93C28)

Cites Work

• Asymptotic Statistics
• Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems
• Title not available (Why is that?)
• A Tutorial on the Positive Realization Problem
• Minimal positive realizations of transfer functions with nonnegative multiple poles
• Title not available (Why is that?)
• Positive realizations of linear systems
• Title not available (Why is that?)
• On the Stability of Switched Positive Linear Systems
• Nonnegative matrix factorization and I-divergence alternating minimization
• Approximation of Nonnegative Systems by Finite Impulse Response Convolutions
• A Note on the Central Limit Theorems for Dependent Random Variables
• A Statistical Model for Positron Emission Tomography
• Order Bound for the Realization of a Combination of Positive Filters
• An algorithm for maximizing expected log investment return
• Deblurring subject to nonnegativity constraints
• Evidence Filtering

Cited In (2)

• Approximation of Nonnegative Systems by Finite Impulse Response Convolutions
• Synchronous deautoconvolution algorithm for discrete-time positive signals via \(\mathcal{I}\)-divergence approximation

This page was built for publication: Approximation of nonnegative systems by moving averages of fixed order

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2280796)