Adaptive covariance estimation of locally stationary processes
Authors:
- Stéphane MallatSMStéphane Mallat
- George PapanicolaouGPGeorge Papanicolaou
- Zhifeng ZhangZZZhifeng Zhang
Open Access
- 1 February 1998
- journal article
- Published by Institute of Mathematical Statistics in The Annals of Statistics
- Vol. 26 (1) , 1-47
Abstract
It is shown that the covariance operator of a locally stationary process has approximate eigenvectors that are local cosine functions. We model locally stationary processes with pseudo-differential operators that are time-varying convolutions. An adaptive covariance estimation is calculated by searching first for a "best" local cosine basis which approximates the covariance by a band or a diagonal matrix. The estimation is obtained from regularized versions of the diagonal coefficients in the best basis.Keywords
This publication has 15 references indexed in Scilit:
- Fitting time series models to nonstationary processesThe Annals of Statistics, 1997
- Wavelet thresholding in anisotropic function classes and application to adaptive estimation of evolutionary spectraThe Annals of Statistics, 1997
- Wavelet Smoothing of Evolutionary Spectra by Nonlinear ThresholdingApplied and Computational Harmonic Analysis, 1996
- Optimal kernels for nonstationary spectral estimationIEEE Transactions on Signal Processing, 1995
- A functional limit theorem for waves reflected by a random mediumApplied Mathematics & Optimization, 1994
- Optimal data-based kernel estimation of evolutionary spectraIEEE Transactions on Signal Processing, 1993
- Wavelets and operators1993
- Entropy-based algorithms for best basis selectionIEEE Transactions on Information Theory, 1992
- Frequency Content of Randomly Scattered SignalsSIAM Review, 1991
- Wigner-Ville spectral analysis of nonstationary processesIEEE Transactions on Acoustics, Speech, and Signal Processing, 1985