A wavelet analysis for time series

From MaRDI portal





In this paper wavelet spectral analysis for a stationary time series is developed [see also, e.g., \textit{M. H. Neumann}, J. Time Ser. Anal. 17, No. 6, 601-633 (1996; Zbl 0873.62099)]. Recently an enormous interest has emerged on the use of wavelets in various areas including statistics. Especially, one can mention signal processing, estimation of spectral densities, changes in hydrologic time series. This paper presents the concept of wavelet spectrum including some examples (with respect to the Haar wavelet). Some asymptotic properties of the finite wavelet transform are proved (e.g. on the asymptotic covariance of the wavelet transform and on asymptotic normality). The wavelet periodogram and its asymptotic properties are discussed. Scalegram at a given scale is defined and illustrated by means of a harmonic time series. Finally, wavelet shrinkage in time series context is presented. Applications to real and simulated series demonstrate the given methodology.



Cites work


Cited in
(42)








This page was built for publication: A wavelet analysis for time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4227977)