Academia.eduAcademia.edu

Outline

Title
Abstract
Introduction
References

Wavelets in time-series analysis

Profile image of Rainer Von SachsRainer Von Sachs

1999, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences

https://doi.org/10.1098/RSTA.1999.0445
visibility

…

description

16 pages

link

1 file

Abstract

This article reviews the role of wavelets in statistical time series analysis. We survey work that emphasises scale such as estimation of variance and the scale exponent of a process with a speci c scale behaviour such as 1/f processes. We present some of our own work on locally stationary wavelet (lsw) processes which model both stationary and some kinds of non-stationary processes. Analysis of time series assuming the lsw model permits identi cation of an evolutionary wavelet spectrum (ews) that quanti es the variation in a time series over a particular scale and at a particular time. We address estimation of the ews and show how our methodology reveals phenomena of interest in an infant electrocardiogram series.

Related papers

Wavelet analysis and scaling properties of time series
Prasanta K. Panigrahi

Physical Review E, 2005

We propose a wavelet based method for the characterization of the scaling behavior of nonstationary time series. It makes use of the built-in ability of the wavelets for capturing the trends in a data set, in variable window sizes. Discrete wavelets from the Daubechies family are used to illustrate the efficacy of this procedure. After studying binomial multifractal time series with the present and earlier approaches of detrending for comparison, we analyze the time series of averaged spin density in the 2D Ising model at the critical temperature, along with several experimental data sets possessing multi-fractal behavior.

View PDFchevron_right
From Fourier to Wavelet Analysis of Time Series
Pedro Morettin

COMPSTAT, 1996

View PDFchevron_right
Analysing multiple time series and extending significance testing in wavelet analysis
Bernard Cazelles

Marine Ecology Progress …, 2008

Following the work of Steele (1985), Pimm & Redfearn (1988) and Lawton (1988), the time–frequency proper-ties of a signal have become of major interest to ecolo-gists (eg Petchey et al. 1997). In nature, non-linear and non-stationary processes are the rule rather than the ...

View PDFchevron_right
Application of Wavelets in Time Series Analysis
sana khan

2011

All praises and thanks be to Almighty, the most beneficent, the most merciful, Who bestowed upon me the courage, patience and strength to embark upon this work and carry it out to its successful completion. After Almighty, I must admire to our last and great Prophet, Huzoor (S.A.W.) whose hfe (as a model) provided me the determination and patience to continue my hard work especially in the period of doubt and despair. Dated: OZ-11' ZOli ^g^Khan) ni Chapter 5 is based on analysis of stock market by using wavelets. In this chapter, we have used wavelet transform to analyze data of stock market. This is one of the application of wavelets in time series analysis. The scatter plots of shares of different companies have been plotted and total variation plots of Sensex versus shares of companies are also shown. At the end, an extensive bibliography of the existing literature related to the subject matter of the dissertation is included.

View PDFchevron_right
High Frequency Time Series Analysis using Wavelets
Jaroslav Schürrer

2015

This paper proposes a new method how to analyze high frequency time series with Wavelets analysis. Our motivation is to introduce a new method, which can be used during decomposition of tick-by-tick stock data for example intraday stock data. Each “tick” is one logical unit of information. High frequency time series are represented by a discrete set of values and usually include tens of thousands of samples per day. In this situation we are dealing with the necessity to reduce input data for further analysis. This paper covers description of high frequency time series, wavelet analysis and method definition for model preparation for analysis

View PDFchevron_right
LOCALLY ADAPTIVE ESTIMATION OF EVOLUTIONARY WAVELET SPECTRA1
Rainer Von Sachs

2000

We introduce a wavelet-based model of local stationarity. This model enlarges the class of locally stationary wavelet processes and contains processes whose spectral density function may change very suddenly in time. A notion of time-varying wavelet spectrum is uniquely defined as a wavelet-type transform of the autocovariance function with respect to so-called autocorrelation wavelets. This leads to a natural representation of the autocovariance which is localized on scales. We propose a pointwise adaptive estimator of the time-varying spectrum. The behavior of the estimator studied in homogeneous and inhomogeneous regions of the wavelet spectrum.

View PDFchevron_right
Consistent Classification of Nonstationary Time Series Using Stochastic Wavelet Representations
Hernando Ombao

Journal of the American Statistical Association, 2009

A method is proposed for classifying an observed non-stationary time series using a bias-corrected non-decimated wavelet transform. Wavelets are ideal for identifying highly discriminant local time and scale features. We view the observed signals as realizations of locally stationary wavelet (LSW) processes. The LSW model provides a timescale decomposition of the signals under which we can define and rigorously estimate the evolutionary wavelet spectrum. The evolutionary spectrum, which contains the second-moment information on the signals, is used as the classification signature. For each time series to be classified, we compute the empirical wavelet spectrum and the divergence from the wavelet spectrum of each group. It is then assigned it to the group to which it is the least dissimilar. Under the LSW framework, we rigorously demonstrate that the classification procedure is consistent, i.e., misclassification probability goes to zero at the rate that is proportional to divergence between the true spectra. The method is illustrated using seismic signals and is demonstrated to work very well in simulation studies.

View PDFchevron_right
A wavelet approach for precursor pattern detection in time series
Russel Pears

Journal of Electrical Systems and Information Technology, 2018

In the context of electrical power systems, identifying precursors to fluctuations in power generation in advance would enable engineers to put in place measures that mitigate against the effects of such fluctuations. In this research we use the Morlet wavelet to transform a time series defined on electrical power generation frequency which was sampled at intervals of 30 s to identify potential precursor patterns. The power spectrum that results is then used to select high coefficient regions that capture a large faction of the energy in the spectrum. We then subjected the high coefficient regions together with a contrasting low coefficient region to a non-parametric ANOVA test and our results indicate that one high coefficient region dominates by predicting an overwhelming percentage of the variation that occurs during the subsequent fluctuation event. These results suggest that the wavelet is an effective mechanism to identify precursor activity in electricity time series data.

View PDFchevron_right
Wavelet: A Technique for Analysis of Ecg
Naveen Dewangan

ijetae.com

The ECG (electrocardiogram) is a very important tool that provides the valuable information about a wide range of cardiac disorders. Medical reports show that the no. of heart patients is increases day-by-day, so it is important to analyze the ECG waveform in an efficient manner. ECG wave commonly change their statistical properties over time, tending to be nonstationary. For analyzing this kind of signal wavelet transforms are a powerful tool. The main tasks in ECG signal analysis are the detection of QRS complex (i.e. R wave), and the estimation of instantaneous heart rate by measuring the time interval between two consecutive R-waves. Wavelet transform provide simultaneous time and frequency information. The wavelet transform decomposes the Electrocardiogram (ECG) signal into a set of frequency band. In the wavelet based algorithm, the ECG signal has been denoised by removing the corresponding wavelet coefficients at higher scales. The analysis has been done on ECG data files of the MIT-BIH Arrhythmia Database.

View PDFchevron_right
Wavelet-based estimators of scaling behavior
J.-f. Muzy, Jean-françois Muzy

IEEE Transactions on Information Theory, 2002

Various wavelet-based estimators of self-similarity or long-range dependence scaling exponent are studied extensively. These estimators mainly include the (bi)orthogonal wavelet estimators and the Wavelet Transform Modulus Maxima (WTMM) estimator. This study focuses both on short and long time-series. In the framework of Fractional Auto-Regressive Integrated Moving Average (FARIMA) processes, we advocate the use of approximately adapted wavelet estimators. For these "ideal" processes, the scaling behavior actually extends down to the smallest scale, i.e., the sampling period of the time series, if an adapted decomposition is used. But in practical situations, there generally exists a cut-off scale below which the scaling behavior no longer holds. We test the robustness of the set of wavelet-based estimators with respect to that cut-off scale as well as to the specific density of the underlying law of the process. In all situations the WTMM estimator is shown to be the best or among the best estimators in terms of the mean square error. We also compare the wavelet estimators with the Detrended Fluctuation Analysis (DFA) estimator which was recently proved to be among the best estimators which are not wavelet-based estimators. The WTMM estimator turns out to be a very competitive estimator which can be further generalized to characterize multiscaling behavior.

View PDFchevron_right

References (33)

  1. Abry, P. Flandrin, P., Taqqu, M. & Veitch, D. 1999 Wavelets for the analysis, estimation and synthesis of scaling data. in: Self Similar Network Tra c Analysis and Performance Evalua- tion. (eds. K. Park and W. Willinger). Wiley.
  2. Abry, P., Gon calv es, P. & Flandrin, P. 1995 Wavelets, spectrum analysis and 1/f processes. In Wavelets and Statistics (ed. A. Antoniadis & G. Oppenheim). Springer Lecture Notes in Statistics, no. 103, pp. 15{29.
  3. Allan, D.W. 1966 Statistics of atomic frequency clocks. Proceedings of the IEEE, 31, 221{230.
  4. Beran, J. 1994 Statistics for long-memory processes. London: Chapman and Hall.
  5. Coifman, R. R. & Donoho, D. L. 1994 Translation-invariant denoising. In Wavelets and Statistics (ed. A. Antoniadis & G. Oppenheim). Springer Lecture Notes in Statistics, no. 103, pp. 125{ 150.
  6. Dahlhaus, R. 1997 Fitting time series models to nonstationary processes. Ann. Statist. 25, 1{37.
  7. Donoho, D. L., Johnstone, I. M., Kerkyacharian, G. & Picard, D. 1995 Wavelet shrinkage: Asymptopia? (with discussion). J. Roy. Statist. Soc. B 57, 301{369.
  8. Farge, M. 1999 Intermittency and coherent votices in fully-developed turbulence. Phil. Trans. R. Soc. Lond. A 357 (to appear).
  9. Flandrin, P. 1998. Time{Frequency / Time{Scale Analysis (Wavelet Analysis and Its Applica- tions, Vol 10). Academic Press, San Diego, CA.
  10. Gao, H.-Ye 1997. Choice of thresholds for wavelet shrinkage estimate of the spectrum. J. Time Series Anal., 18, 231{251.
  11. Gon calv es, P. & Flandrin, P. 1993 Bilinear Time-Scale Analysis Applied to Local Scaling Ex- ponents Estimation. in: Progress in Wavelet Analysis and Applications (Y. Meyer and S. Roques, eds.), pp. 271{276, Editions Frontieres, Gif-sur-Yvette.
  12. Johnstone, I.M. 1999 Wavelets and the theory of nonparametric function estimation. Phil. Trans. R. Soc. Lond. A 357 (to appear).
  13. Ho mann, M. 1999. On nonparametric estimation in nonlinear AR(1)-models. Statistics and probability letters (to appear).
  14. McCoy, E.J. & Walden, A.T. 1996 Wavelet analysis and synthesis of stationary long-memory processes. Journal of Computational and Graphical Statistics, 5, 26{56.
  15. Nason, G.P., Sapatinas, T. & Sawczenko, A. 1999 Statistical modelling of time series using non-decimated wavelet representations. Preprint, University of Bristol.
  16. Nason, G. P. & Silverman, B. W. 1995 The stationary wavelet transform and some statisti- cal applications. In Wavelets and Statistics (ed. A. Antoniadis & G. Oppenheim). Springer Lecture Notes in Statistics, no. 103, pp. 281{300.
  17. Nason, G.P., von Sachs, R. & Kroisandt, G. 1998 Wavelet processes and adaptive estimation of the evolutionary wavelet spectrum. Discussion Paper 98/22, Institut de Statistique, UCL, Louvain-la-Neuve.
  18. Neumann, M. 1996. Spectral density estimation via nonlinear wavelet methods for stationary non-Gaussian time series. J. Time Ser. Anal., 17, 601{633.
  19. Neumann, M. & von Sachs, R. 1997. Wavelet thresholding in anisotropic function classes and application to adaptive estimation of evolutionary spectra. Ann. Statist., 25, 38{76.
  20. Page, C.H. 1952 Instantaneous power spectra. J. Appl. Phys. , 23, 103{106.
  21. Percival, D.B. & Guttorp, P. 1994 Long-memory processes, the Allan variance and wavelets. In Wavelets in Geophysics (ed. E. Foufoula-Georgiou & P. Kumar), pp. 325{344, Academic Press.
  22. Priestley, M.B. 1965 Evolutionary spectra and non-stationary processes. J. Roy. Statist. Soc. B 27, 204{237.
  23. Priestley, M.B. 1981 Spectral Analysis and Time Series. London: Academic Press.
  24. Rioul, O. & Vetterli, M. 1991 Wavelets and signal processing. IEEE Sig. Proc. Mag., 8, 14{38.
  25. Serroukh, A., Walden, A.T. & Percival, D.B. 1998 Statistical properties of the wavelet variance estimator for non-Gaussian/non-linear time series. Statistics Section Technical Report, TR- 98-03. Department of Mathematics, Imperial College, London.
  26. von Sachs, R. & MacGibbon B. 1998 Nonparametric curve estimation by wavelet thresholding with locally stationary errors (submitted for publication).
  27. von Sachs, R. & Neumann, M. 1998 A wavelet-based test for stationarity (submitted for publi- cation).
  28. Sachs, R. & Schneider, K. 1996 Wavelet smoothing of evolutionary spectra by non-linear thresholding. Appl Comp. Harmonic Anal. , 3, 268{282.
  29. Silverman, B.W. 1999 Wavelets in statistics: beyond the standard assumptions. Phil. Trans. R. Soc. Lond. A 357 (to appear).
  30. Silverman, R.A. 1957 Locally stationary random processes. IRE Trans. Information Theory, IT-3, 182{187.
  31. Wang, Y., Cavanaugh, J. and Song, Ch. 1997 Self-similarity index estimation via wavelets for locally self-similar processes. Preprint, Dept. Stat., University of Missouri.
  32. Whitcher, B., Byers, S.D., Guttorp, P. & Percival, D.B. 1998 Testing for homogeneity of variance in time series: long memory, wavelets and the Nile river. Submitted for publication.
  33. Wornell, G.W. and Oppenheim, A.V. 1992 Estimation of fractal signals from noisy measurements using wavelets. IEEE Transactions on signal processing, 40, 611{623.

Related papers

Wavelet processes and adaptive estimation of the evolutionary wavelet spectrum
Rainer Von Sachs

Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2000

This article defines and studies a new class of non-stationary random processes constructed from discrete non-decimated wavelets which generalizes the Cramér (Fourier) representation of stationary time series. We define an evolutionary wavelet spectrum (EWS) which quantifies how process power varies locally over time and scale. We show how the EWS may be rigorously estimated by a smoothed wavelet periodogram and how both these quantities may be inverted to provide an estimable time-localized autocovariance. We illustrate our theory with a pedagogical example based on discrete non-decimated Haar wavelets and also a real medical time series example.

View PDFchevron_right
A wavelet analysis for time series
Pedro Morettin

Journal of Nonparametric Statistics, 1998

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

View PDFchevron_right
Some aspects of wavelet analysis in time series
Abdullah Almasri

2000

This thesis consists of two papers dealing with the methodology of wavelet and its application in time series. We give here a brief summary of the contents of the two papers in the thesis. The first paper describes an alternative approach for testing the existence of trend among time series. The test method has been constructed using wavelet analysis which have the ability to decompose a time series into low frequencies (trend) and high frequencies (noise) components. Under the normality assumption the test is distributed as F. However, the distribution of the test is unknown under other conditions, like non-normality. To investigate the properties of the test statistic under

View PDFchevron_right
Wavelets for time series analysis - a survey and new results
Piotr Wojdyllo

Control and Cybernetics, 2005

In the paper we review stochastic properties of wavelet coefficients for time series indexed by continuous or discrete time. The main emphasis is on decorrelation property and its implications for data analysis. Some new properties are developed as the rates of correlation decay for the wavelet coefficients in the case of longrange dependent processes such as the fractional Gaussian noise and the fractional autoregressive integrated moving average processes. It is proved that for such processes the within-scale covariance of the wavelet coefficients at lag k is O(k 2(H−N)−2), where H is the Hurst exponent and N is the number of vanishing moments of the wavelet employed. Some applications of decorrelation property are briefly discussed.

View PDFchevron_right
Wavelets in Statistics
Pedro Morettin

Resenhas do Instituto de Matemática e Estatística da Universidade de São Paulo, 1997

In this paper we give the main uses of wavelets in statistics, with emphasis in time series analysis. We include the fundamental work on non parametric regression, which motivated the development of techniques used in the estimation of the spectral density of stationary processes and of the evolutionary spectrum of locally stationary processes. We also mention briefly some further topics, as density estimation and Bayesian analysis.

View PDFchevron_right

Cited by

A survey on wavelet applications in data mining
Mitsunori Ogihara

ACM SIGKDD Explorations Newsletter, 2002

Recently there has been significant development in the use of wavelet methods in various data mining processes. However, there has been written no comprehensive survey available on the topic. The goal of this is paper to fill the void. First, the paper presents a high-level data-mining framework that reduces the overall process into smaller components. Then applications of wavelets for each component are reviewd. The paper concludes by discussing the impact of wavelets on data mining research and outlining potential future research directions and applications.

View PDFchevron_right
Forecasting monthly precipitation in Central Chile: a self-organizing map approach using filtered sea surface temperature
Max Billib, José Arumí, Cintia Bertacchi Uvo

Theoretical and Applied Climatology, 2012

Western South America is subject to considerable inter-annual variability due to El Niño-Southern Oscillation (ENSO) so forecasting inter-annual variations associated with ENSO would provide an opportunity to tailor management decisions more appropriately to the season. On one hand, the self-organizing maps (SOM) method is a suitable technique to explore the association between sea surface temperature and precipitation fields. On the other hand, Wavelet transform is a filtering technique, which allows the identification of relevant frequencies in signals, and also allows localization on time. Taking advantage of both methods, we present a method to forecast monthly precipitation using the SOM trained with filtered SST anomalies. The use of the SOM to forecast precipitation for Chillan showed good agreement between forecasted and measured values, with correlation coefficients (r 2 ) ranging from 0.72 to 0.91, making the combined use filtered SST fields and SOM a suitable tool to assist water management, for example in agricultural water management. The method can be easily tailored to be applied in other stations or to other variables.

View PDFchevron_right
Low and high flow analyses and wavelet application for characterization of the Blue Nile River system
Tibebe Dessalegne, Wossenu Abtew, Assefa M Melesse

Hydrological Processes, 2009

The low and high flow characteristic of the Blue Nile River (BNR) basin is presented. The study discusses low and high flow, flow duration curve (FDC) and trend analysis of the BNR and its major tributaries. Different probability density functions were fitted to better describe the low and high flows of the BNR and major tributaries in the basin. Wavelet analysis was used in understanding the variance and frequency-time localization and detection of dominant oscillations in rainfall and flow. FDCs were developed, and low flow (below 50% exceedance) and high flow (over 75% exceedance) of the curves were analysed and compared. The Gravity Recovery and Climate Experiment (GRACE) satellite-based maps of monthly changes in gravity converted to water equivalents from 2003 to 2006 for February, May and September showed an increase in the moisture influx in the BNR basin for the month of September, and loss of moisture in February and May. It was also shown that 2004 and 2005 were drier with less moisture influx compared to 2003 and 2006. On the basis of the Kolmogorov-Smirnov, Anderson-Darling and Chi-square tests, Gen. Pareto, Frechet 3P, Log-normal, Log-logistics, Fatigue Life and Phased Bi-Weibull distributions best describe the low and high flows within the BNR basin. This will be beneficial in developing flow hydrographs for similar ungauged watersheds within the BNR basin. The below 50% and above 75% exceedance on the FDC for five major rivers in addition to the BNR showed different characteristics depending on size, land cover, topography and other factors. The low flow frequency analysis of the BNR at Bahir Dar showed 0Ð55 m 3 /s as the monthly low flow with recurrence interval of 10 years. The wavelet analysis of the rainfall (at Bahir Dar and basin-wide) and flows at three selected stations shows interand intra-annual variability of rainfall and flows at various scales.

View PDFchevron_right
Case study: shipping trend estimation and prediction via multiscale variance stabilisation
Antonis Michis

Journal of Applied Statistics, 2016

Shipping and shipping services are a key industry of great importance to the economy of Cyprus and the wider European Union. Assessment, management and future steering of the industry, and its associated economy, is carried out by a range of organisations and is of direct interest to a number of stakeholders. This article presents an analysis of shipping credit flow data: an important and archetypal series whose analysis is hampered by rapid changes of variance. Our analysis uses the recently developed data-driven Haar-Fisz transformation that enables accurate trend estimation and successful prediction in these kinds of situation. Our trend estimation is augmented by bootstrap confidence bands, new in this context. The good performance of the data-driven Haar-Fisz transform contrasts with the poor performance exhibited by popular and established variance stabilisation alternatives: the Box-Cox, logarithm and square root transformations.

View PDFchevron_right
Trends of Runoff Variation and Effects of Main Causal Factors in Mun River, Thailand During 1980–2018
Arika Bridhikitti

Water

Mun River is the largest tributary of the Mekong River in Thailand and provides abundant water resources not only for an important agricultural area in Thailand but also for the lower Mekong River. To understand how the runoff of Mun River responds to climate change and human activities in recent decades, this study performed a detailed examination of the characteristics of runoff variation based on measurements at two hydrological gauging stations on the main stem of Mun River during 1980–2018. Using the Mann-Kendall test, Morlet wavelet transform and Double Cumulative Curve methods, this study identifies that the variation of annual runoff of Mun River encountered an abruption in 1999/2000, with an increased trend taking place since then. Furthermore, a detailed assessment of the effects of the variations in rainfall, temperature, evaporation, and land use types extracted from remote sensing images at the basin scale reveals that a significant reduction in forest area and slight r...

View PDFchevron_right
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved