Optimization with respect to covariance sequence parameters

DOI10.1016/0005-1098(85)90041-XMaRDI QIDQ1070991zbMATH OpenOpenAlexFDO

Authors Torsten Söderström, Petre Stoica

Publication date 1985

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0005-1098(85)90041-x

zbMATH Keywords

partial autocorrelation functionrecursive algorithmsstationary discrete process

Mathematics Subject Classification ID

Signal detection and filtering (aspects of stochastic processes) (60G35) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Estimation and detection in stochastic control theory (93E10) Optimal stochastic control (93E20)

Review text

Let \(\rho_ k\) denote the covariance function of some stationary discrete process and let \(\rho =(\rho_ 0,...,\rho_ p)'\), \(p\geq 1\). Consider the problem of optimizing some scalar function f(\(\rho)\). The main difficulty associated with this problem is that the covariance sequence must be nonnegative definite. The paper presents a procedure which overcomes this difficulty by optimizing the problem in terms of the partial autocorrelation function.

Recommendations

• scientific article; zbMATH DE number 3921781
• Stationary discrete-time covariance factorization using Newton-Raphson iteration
• Optimal nonnegative definite approximations of estimated moving average covariance sequences
• Extrapolation problem for functionals from a periodically correlated sequence
• scientific article; zbMATH DE number 3911535

Cites work

• A useful input parameterization for optimal experiment design
• Characterization of the partial autocorrelation function
• scientific article; zbMATH DE number 44406 (Why is no real title available?)
• scientific article; zbMATH DE number 3465192 (Why is no real title available?)
• scientific article; zbMATH DE number 3565994 (Why is no real title available?)
• Instrumental variable methods for system identification
• Maximum Likelihood Fitting of ARMA Models to Time Series with Missing Observations
• The Fitting of Time-Series Models

Cited in

(8)

• Optimising prediction error among completely monotone covariance sequences
• Stationary discrete-time covariance factorization using Newton-Raphson iteration
• Approximate maximum likelihood frequency estimation
• Optimization of parameters of stationary models in a unitary space
• Approximate maximum-likelihood approach to ARMA spectral estimation
• scientific article; zbMATH DE number 4043641 (Why is no real title available?)
• A generalised instrumental variable estimator for multivariable errors-in-variables identification problems
• Optimization of the generalized covariance estimator in noncausal processes

This page was built for publication: Optimization with respect to covariance sequence parameters

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1070991)