Wavelet-based Forecasting of Short and Long Memory Time

A wavelet-based forecasting method for time series is introduced. It is based on a multiple resolution decomposition of the signal, using the redundant “a trous” wavelet transform which has the advantage of being shift-invariant. The result is a decomposition of the signal into a range of frequency scales. The prediction is based on a small number of coefficients on each of these scales. In its simplest form it is a linear prediction based on a wavelet transform of the signal. This method uses sparse modelling, but can be based on coefficients that are summaries or characteristics of large parts of the signal. The lower level of the decomposition can capture the longrange dependencies with only a few coefficients, while the higher levels capture the usual short-term dependencies. We show the convergence of the method towards the optimal prediction in the autoregressive case. The method works well, as shown in simulation studies, and studies involving financial data. Index Terms Wavelet transform, forecasting, resolution, scale, autoregression, time series, model

[1]  Stéphane Mallat,et al.  Analysis of low bit rate image transform coding , 1998, IEEE Trans. Signal Process..

[2]  A. Walden,et al.  Wavelet Analysis and Synthesis of Stationary Long-Memory Processes , 1996 .

[3]  C. Chui.  Wavelet Analysis and Its Applications , 1992 .

[4]  Jiao Licheng,et al.  Wavelet model for the time scale , 1999, Proceedings of the 1999 Joint Meeting of the European Frequency and Time Forum and the IEEE International Frequency Control Symposium (Cat. No.99CH36313).

[5]  M.E. El-Hawary,et al.  Short term load forecasting by using wavelet neural networks , 2000, 2000 Canadian Conference on Electrical and Computer Engineering. Conference Proceedings. Navigating to a New Era (Cat. No.00TH8492).

[6]  David S. Stoffer,et al.  Time series analysis and its applications , 2000 .

[7]  L. Hong,et al.  A filter-bank-based Kalman filtering technique for wavelet estimation and decomposition of random signals , 1998 .

[8]  Israel Cohen,et al.  Orthonormal shift-invariant wavelet packet decomposition and representation , 1997, Signal Process..

[9]  Fionn Murtagh,et al.  Image Processing and Data Analysis - The Multiscale Approach , 1998 .

[10]  D. Donoho,et al.  Translation-Invariant De-Noising , 1995 .

[11]  S. Elangovan,et al.  Applications of symlets for denoising and load forecasting , 1999, Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics. SPW-HOS '99.

[12]  Nouna Kettaneh,et al.  Statistical Modeling by Wavelets , 1999, Technometrics.

[13]  Richard A. Davis,et al.  Time Series: Theory and Methods , 2013 .

[14]  Todd R. Ogden,et al.  Wavelet Methods for Time Series Analysis , 2002 .

[15]  Lizhong Wu,et al.  What is the "true price"? state space models for high frequency FX data , 1997, Proceedings of the IEEE/IAFE 1997 Computational Intelligence for Financial Engineering (CIFEr).

[16]  Murali Tummala,et al.  Multirate, multiresolution, recursive Kalman filter , 2000, Signal Process..

[17]  Mark J. Shensa,et al.  The discrete wavelet transform: wedding the a trous and Mallat algorithms , 1992, IEEE Trans. Signal Process..

[18]  P. Dutilleux.  An Implementation of the “algorithme à trous” to Compute the Wavelet Transform , 1989 .

[19]  Alan S. Willsky,et al.  Multiscale Autoregressive Models and Wavelets , 1999, IEEE Trans. Inf. Theory.

[20]  Ingrid Daubechies,et al.  Ten Lectures on Wavelets , 1992 .

[21]  A. Walden,et al.  Wavelet Methods for Time Series Analysis , 2000 .

[22]  Stéphane Canu,et al.  The long-term memory prediction by multiscale decomposition , 2000, Signal Process..

[23]  V. Bjorn.  Multiresolution methods for financial time series prediction , 1995, Proceedings of 1995 Conference on Computational Intelligence for Financial Engineering (CIFEr).