Time series analysis with wavelet coefficients



Time series are conventionally analyzed only in the time domain or only in the frequency domain, and few analyses make use of information in both domains simultaneously. On the other hand, time series analysis based on the wavelet transform has been concentrated on the irregularity detection or the analysis of stochastic processes constructed by the wavelet transform. The wavelet transform is applied to stationarity analysis and predictions in the present paper. Using the wavelet transform, we can decompose time series into frequency components. Consequently, we can extract local information with respect to frequency. We observe the time series in both the time domain and the frequency domain simultaneously. And we connect weak stationarity and prediction methods of original time series to those of each frequency component, accompanied with numeric results.

Key words

wavelet transform stationary wavelet transform stochastic processes weak stationarity test predictions 


  1. [1]
    D. Allingham, M. West and A.I. Mees, Wavelet reconstruction of nonlinear dynamics. Int. J. of Bifurcation and Chaos,8 (1998), 2191–2201.MATHCrossRefGoogle Scholar
  2. [2]
    L. Cao, Y. Hong, H. Fang and G. He, Predicting chaotic time series with wavelet networks. Physica,D85 (1995), 225–238.Google Scholar
  3. [3]
    M. Casdagli, Nonlinear prediction of chaotic time series. Physica,D35 (1989), 335–356.MathSciNetGoogle Scholar
  4. [4]
    R. Christensen, Linear Models for Multivariate Time Series, and Spatial Data. Springer- Verlag, New York, 1991.MATHGoogle Scholar
  5. [5]
    Charles K. Chui, An Introduction to Wavelets. Academic Press, 1992.Google Scholar
  6. [6]
    Charles K. Chui, Wavelets: A Mathematical Tool for Signal Analysis. Philadelphia Society for Industrial and Applied Mathematics, 1997.Google Scholar
  7. [7]
    E.J. Hannan, P.R. Krishnaiah and M.M. Rao (eds.), Handbook of Statistics 5: Time Series in the Time Domain. Elsevier Science Publishers B. V., Amsterdam, 1985.Google Scholar
  8. [8]
    A.C. Harvey, Time Series Models. Philip Allan, Oxford, 1981.MATHGoogle Scholar
  9. [9]
    S. Ihara, Stochastic Processes and Entropy. Iwanami, 1984 (in Japanese).Google Scholar
  10. [10]
    T. Kawada, Stationary Stochastic Process. Kyoritsu, 1985 (in Japanese).Google Scholar
  11. [11]
    N. Masuda and K. Aihara, The prediction of chaotic time series with wavelet coefficients. IEICE,J82-A, No. 11 (1999), 1710–1718 (in Japanese). The English version is to appear in Electronics and Communications in Japan Part III: Fundamental Electronic Science.Google Scholar
  12. [12]
    G.P. Nason and B.W. Silverman, The stationary wavelet transform and some statistical applications. Wavelet and Statistics (eds. A. Antoniadis and G. Oppenheim), Lecture Notes in Statistics103, Springer-Verlag, New York, 1995, 281–299.Google Scholar
  13. [13]
    G.P. Nason, T. Sapatinas and A. Sawczenko, Statistical modeling of time series using nondecimated wavelet representations. Technical Report, Dep. of Math., University of Bristol, Bristol, 1997.Google Scholar
  14. [14]
    Y. Okabe and Y. Nakano, The theory of KM2O-Langevin equations and its applications to data analysis (I): Stationary analysis. Hokkaido Math. J.,20 (1991), 45–90.MATHMathSciNetGoogle Scholar
  15. [15]
    Y. Okabe and A. Inoue, The theory of KM2O-Langevin equations and its applications to data analysis (II): Causal analysis (I). Nagoya Math. J.,134 (1994), 1–28.MATHMathSciNetGoogle Scholar
  16. [16]
    Y. Okabe and T. Yamane, The theory of KM2O-Langevin equations and its applications to data analysis (III): Deterministic analysis. Nagoya Math. J.,152 (1998), 175–201.MATHMathSciNetGoogle Scholar
  17. [17]
    R.von Sachs, G.P. Nason and G. Kroisandt, Adaptive estimation of the evolutionary wavelet spectrum. Technical Report 516, Dep. of Stat., Stanford University, Stanford, CA, 1997.Google Scholar
  18. [18]
    M. Satou, Mathematical foundation of wavelet theory. J. Acous. of Japan,47, No. 6 (1991), 406–423 (in Japanese).Google Scholar
  19. [19]
    W. Vandaele, Applied Time Series and Box-Jenkins Models. Academic Press, New York, 1983.MATHGoogle Scholar

Copyright information

© JJIAM Publishing Committee 2001

Authors and Affiliations

  1. 1.Department of Mathematical Engineering and Information Physics, Graduate School and Faculty of EngineeringUniversity of TokyoTokyoJapan

Personalised recommendations