Embedding nonnegative definite Toeplitz matrices in nonnegative definite circulant matrices, with application to covariance estimation

DOI10.1109/18.45276zbMath0698.62088OpenAlexW2115125420MaRDI QIDQ3476160

Colin L. Mallows, Lawrence A. Shepp, Amir Dembo

Publication date: 1989

Published in: IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL: https://semanticscholar.org/paper/c144448bc1266d8c6564da6e81af741c968ebe81

zbMATH Keywords

stationary Gaussian processmaximum likelihood estimateextremal processesminimal eigenvaluecovariance of a periodic stationary processEM iterative algorithmfinite duration sequencenonnegative definite circulant matricesnonnegative definite Toeplitz matrices

Mathematics Subject Classification ID

Non-Markovian processes: estimation (62M09) Inference from stochastic processes (62M99)

Related Items

Every matrix is a product of Toeplitz matrices, Realisability conditions for second-order marginals of biphased media, When is a truncated covariance function on the line a covariance function on the circle?, Representations of Gaussian random fields and approximation of elliptic PDEs with lognormal coefficients, A regularised estimator for long-range dependent processes, Periodic vector processes with an internal reciprocal dynamics, An efficient algorithm for maximum entropy extension of block-circulant covariance matrices, Fast and exact synthesis of stationary multivariate Gaussian time series using circulant embedding, A maximum entropy approach to the realizability of spin correlation matrices, Unnamed Item