Wavelets in statistics: A review

  • A. Antoniadis


The field of nonparametric function estimation has broadened its appeal in recent years with an array of new tools for statistical analysis. In particular, theoretical and applied research on the field of wavelets has had noticeable influence on statistical topics such as nonparametric regression, nonparametric density estimation, nonparametric discrimination and many other related topics. This is a survey article that attempts to synthetize a broad variety of work on wavelets in statistics and includes some recent developments in nonparametric curve estimation that have been omitted from review articles and books on the subject. After a short introduction to wavelet theory, wavelets are treated in the familiar context of estimation of «smooth» functions. Both «linear» and «nonlinear» wavelet estimation methods are discussed and cross-validation methods for choosing the smoothing parameters are addressed. Finally, some areas of related research are mentioned, such as hypothesis testing, model selection, hazard rate estimation for censored data, and nonparametric change-point problems. The closing section formulates some promising research directions relating to wavelets in statistics.

Keywords and phrares

Wavelets multiresolution analysis nonparametric curve estimation density estimation regression model selection orthogonal series thresholding crowss-validation shrinkage denoising 


  1. [1]
    Abramovich, F., Sapatinas, T. andSilverman, B. W. (1998). Wavelet thresholding via a Bayesian approach. I. Royal Statist. Soc., Ser. B, part 4, 725–750.CrossRefMathSciNetGoogle Scholar
  2. [2]
    Amato, U. andVuza, D. T. (1994). Wavelet Regularization for Smoothing Data. Technical report N. 108/94, Instituto per Applicazioni della Matematica, Napoli.Google Scholar
  3. [3]
    Amato, U. andVuza, D. T. (1996). An Alternate Proof of a Result of Johnstone and Silverman concering Wavelet Thresholding Estimators for Data with Correlated Noise.Revue Roumaine Math. Pures Appi. 41, 431–438.zbMATHMathSciNetGoogle Scholar
  4. [4]
    Antoniadis, A. (1994). Smoothing noisy data with coiflets.Statistica Sinica 4 (2), 651–678.zbMATHMathSciNetGoogle Scholar
  5. [5]
    Antoniadis, A. (1994). Smoothing noisy data with tapered coiflets series.Scand. Journal of Statistics 23, 313–330.MathSciNetGoogle Scholar
  6. [6]
    Antoniadis, A. andCarmona, R. (1991). Multiresolution analyses and wavelets for density estimation. Technical report, University of California, Irvine.Google Scholar
  7. [7]
    Antoniadis, A., Grégoire, G. andMcKeague, I. (1994). Wavelet methods for curve estimation.J. Amer Statist. Assoc. 89 (428), 1340–1353.zbMATHCrossRefMathSciNetGoogle Scholar
  8. [8]
    Antoniadis, A. andPham, D. T. (1995). Wavelet regression for random or irregular design. Technical report. University of Grenoble.Google Scholar
  9. [9]
    Antoniadis, A. andLavergne, C. (1995). Variance function estimation in regression with wavelet methods. In A. Antoniadis and G. Oppenheim (eds.),Wavelets and Statistics. Lecture Notes in Statistics,103, Springer-Verlag.Google Scholar
  10. [10]
    Antoniadis, A., Grégoire, G. andVial, P. (1997a). Random design wavelet curve smoothing.Statistics and Prob. Letters, vol. 35, 225–232.zbMATHCrossRefGoogle Scholar
  11. [11]
    Antoniadis, A., Gijbels, I. andGrégotre, G. (1997b). Model selection using wavelet decomposition and applications. Biometrika84 (4), 751–763.zbMATHCrossRefMathSciNetGoogle Scholar
  12. [12]
    Antoniadis, A., Grégoire, G. andNason, G. (1997c). Density and hazard rate estimation for right censored data using wavelet methods. Technical report. University of Grenoble.Google Scholar
  13. [13]
    Antoniadis, A. andGijbels, I. (1997). Detecting abrupt changes by wavelet methods. Technical report. University of Grenoble.Google Scholar
  14. [14]
    Averkamp, R. andHoudré, C. (1996). Wavelet thresholding for non (necessarily) Gaussian noise: a preliminary report. Technical report, Georgia Institute of Technology, Atlanta.Google Scholar
  15. [15]
    Beran, R. (1994). Bootsrap variable selection and confidence sets. Technical report, University of California, Berkeley.Google Scholar
  16. [16]
    Brillinger, D. R. (1994). Some River Wavelets.Environmetrics 5, 2 11–220.Google Scholar
  17. [17]
    Brillinger, D.R. (1995). Some Uses of Cumulants in Wvelet Analysis.J. Nonparam. Statistics 4.Google Scholar
  18. [18]
    Bruce, A. G. andGao, H.-Y. (1194).S + Wavelets, Users manual. StatSci, Seatle.Google Scholar
  19. [19]
    Bruce, A. G. andGao, H.-Y. (1996). Understanding WaveShrink: Variance and Bias Estimation. Biometrika83 (4), 727–746.zbMATHCrossRefMathSciNetGoogle Scholar
  20. [20]
    Bruce, A. G. andGao, H.-Y. (1997). WaveShink with firm shrinkage.Statistica Sinica, to appear.Google Scholar
  21. [21]
    Buckhett, J. B. andDonoho, D. (1995). Wavelab and Reproducible research. In A. Antoniadis and G. Oppenheim (eds.),Wavelet and Statistics, Lecture Notes in Statistics,103, Springer-Verlag.Google Scholar
  22. [22]
    Clyde, M. Parmigiani, G. andVidakovic, B. (1995). Multiple shrinkage and subset selection in wavelets. Technical report DP 95-37, Duke University.Google Scholar
  23. [23]
    Chipman, H. A., Kolaczyj, E. D. andMcculloch, R. E. (1995). Adaptive Bayesian Wavelet Shrinkage. Technical report, University of Chicago.Google Scholar
  24. [24]
    Chui, K. (1992).Wavelets: A Tutorial in Theory and Applications. Academic Press, Boston.zbMATHGoogle Scholar
  25. [25]
    Cohen, A., Daubechies, I. andVial, P. (1993). Wavelets on the interval and fast wavelet transforms.Applied and Comp. Harmonic Analysis 1 (1), 54–81.zbMATHCrossRefMathSciNetGoogle Scholar
  26. [26]
    Cohen A. andRyan, R. D. (1995).Wavelets and Multiscale Signal Processing. Chapman & Hall, London.zbMATHGoogle Scholar
  27. [27]
    Daubechies, I. (1992).Ten Lectures of Wavelets. CBMS-NSF regional conferences series in applied mathematics. SIAM, Philadelphia.Google Scholar
  28. [28]
    Daubechies, I. andLagarias, J. C. (1991). Two-scale difference equations: Existence and global regularity of solutions.SIAM Journal on Math. Analysis 22, 1388–1410.zbMATHCrossRefMathSciNetGoogle Scholar
  29. [29]
    Delyon, B. andJuditsky, A. (1996). On minimax wavelet estimators.Applied and Comp. Harmonic Analysis 3, 215–228.zbMATHCrossRefMathSciNetGoogle Scholar
  30. [30]
    DeVore, R. andLucier, B. J. (1992). Fast wavelet techniques for near-optimal signal processing. InIEEE Military Communication Conference, pp. 1129–1135.Google Scholar
  31. [31]
    DeVore, R., Jawerth, B. andPoov, V. (1988). Interpolation of Besov spaces.Trans. Amer Math. Soc. 305, 397–414.zbMATHCrossRefMathSciNetGoogle Scholar
  32. [32]
    Donoho, D. L. (1994). Asymptotic risk for sup-norm loss: solution via optimal recovery. Prob.Theory and Related Fields 99, 145–170.zbMATHCrossRefMathSciNetGoogle Scholar
  33. [33]
    Donoho, D. L. andJohnstone, I. M. (1992). Minimax estimation via wavelet shrinkage. Technical report, Stanford University.Google Scholar
  34. [34]
    Donoho, D. L. andJohnstone, I. M. (1994a). Ideal spatial adaptation by wavelet shrinkage.Biometrika 81, 425–455.zbMATHCrossRefMathSciNetGoogle Scholar
  35. [35]
    Donoho, D. L. andJohnstone, I. M. (1995). Adapting to unknown smoothness via wavelet shrinking.J. Am. Statist. Assoc. 90, 1200–1224.zbMATHCrossRefMathSciNetGoogle Scholar
  36. [36]
    Donoho, D. L. andJohnstone, I. M. (1994b). Ideal denoising in an orthonormal basis chosen from a library of bases.Compt. Rend. Acad. Sci. Paris A 319, 1317–1322.zbMATHMathSciNetGoogle Scholar
  37. [37]
    Donoho, D. L., Johnstone, I. M., Kerkyacharian, G. andPicard, D. (1995). Wavelet shrinkage: asymptotia (with discussion)?J. Roy. Statist. Soc., Ser. B 57(2), 301–370.zbMATHMathSciNetGoogle Scholar
  38. [38]
    Donoho, D. L., Johnstone, I. M., Kerkyacharian, G. andPicard, D. (1996). Density estimation by wavelet thresholding.Ann. Statist. 24 (2), 508–539.zbMATHCrossRefMathSciNetGoogle Scholar
  39. [39]
    Doukhan, P. (1990). Consistency of delta-sequence estimates of a density or of a regression function for a weakly dependent stationary sequence. Séminaire de Statistique d’Orsay, Université Paris Sud, 1991.Google Scholar
  40. [40]
    Doukhan, P. andLéon, J. (1990). Déviation quadratique d’estimateurs de densité par projection orthogonale.Compt. Rend. Acad. Sci. Paris A 310, 424–430.Google Scholar
  41. [41]
    Engel, J. (1990). Density estimation with Haar series.Statistics and Probability Letters 9, 111–117.zbMATHCrossRefMathSciNetGoogle Scholar
  42. [42]
    Fan, J. (1996). Test of significance based on wavelet thresholding and Neyman’s truncation.J. Am. Statist. Assoc. 91, 674–688.zbMATHCrossRefGoogle Scholar
  43. [43]
    Gao, H.-Y. (1993).Wavelet estimation of Spectral densities in Time series analysis. Ph. D. Thesis, University of California, Berkeley.Google Scholar
  44. [44]
    Gao, H.-Y. (1997). Wavelet Shrinkage Smoothing For Heteroscedastic Data. Technical report, ScatSci, Seatle.Google Scholar
  45. [45]
    Gasser, T. andMüller, H. (1979). Kernel estimation of regression functions. In Gasser, T. and Müller, H. (eds.),Curve Estimation, Springer-Verlag, Heidelberg.CrossRefGoogle Scholar
  46. [46]
    Good, I. J. andGaskins, R. A. (1971). Density estimation and bump haunting by the penalized maximum likelihood method.J. Am. Statist. Assoc. 75, 42–69.CrossRefMathSciNetGoogle Scholar
  47. [47]
    Hall, O. andNason, G. P. (1996). On choosing a non-integer resolution level when using wavelet methods. Technical report, University of Bristol.Google Scholar
  48. [48]
    Hall, P. andPatil, P. (1995). On wavelet methods for estimating smooth functions.Bernoulli 1, 41–58.zbMATHCrossRefMathSciNetGoogle Scholar
  49. [49]
    Hall, P. andPatil, P. (1996a). Effect of threshold rules on performance of wavelet-based curve estimators.Statistica Sinica 6, 331–345.zbMATHMathSciNetGoogle Scholar
  50. [50]
    Hall, P. andPatil, P. (1996b). On the choice of smoothing parameter, threshold and truncation in nonparametric regression by nonlinear wavelet methods.J. Roy. Statist. Soc., Ser B 58, 361–377.zbMATHMathSciNetGoogle Scholar
  51. [51]
    Härdle, W. (1990).Applied nonparametric regression. Cambridge University Press, Cambridge.zbMATHGoogle Scholar
  52. [52]
    Holschneider, M. (1995).Wavelets: An analysis tool. Clarendon Press, Oxford.zbMATHGoogle Scholar
  53. [53]
    Jansen, M., Malfait, M. andBultheel, A. (1997). Generalized cross-validation for wavelet thresholding.Signal Processing,56 (1). To appear.Google Scholar
  54. [54]
    Johnstone, I. M., Kerkyacharian, G. andPicard, D. (1992). Estimation d’une densité de probabilityé par méthode d’ondelettes.Compt. Rend. Acad. Sci. Paris A 315, 211–216.zbMATHMathSciNetGoogle Scholar
  55. [55]
    Johnstone, I. M. andSilverman, B. W. (1997). Wavelet threshold estimators for data with correlated noise.J. Roy. Statist. Soc., Ser B 59. In press.Google Scholar
  56. [56]
    Kerkyacharian, G. andPicard, D. (1992). Density estimation in Besov Spaces.Statistics and Probability Letters 13, 15–24.zbMATHCrossRefMathSciNetGoogle Scholar
  57. [57]
    Kolaczyk, E. (1994).Wavelet methods for the inversion of some homogeneous linear operators in the presence of noisy data. Ph. D. Thesis, Stanford University.Google Scholar
  58. [58]
    Mallat, S. G. (1989). A theory for multiresolution signal decomposition: the wavelet representation.IEEE Trans. on Pattern Analysis and Machine Intelligence 11, 674–693.zbMATHCrossRefGoogle Scholar
  59. [59]
    Marron, S. J., Adak, S., Johnstone, I., Neumann, M. andPatil, P. (1997). Exact risk analysis of wavelet regression.Journal of Computational and Graphical Statistics. In press.Google Scholar
  60. [60]
    Masry, E. (1994). Probability density estimation from dependent observations using wavelet orthonormal bases.Statistics and Probability Letters 21, 181–194.zbMATHCrossRefMathSciNetGoogle Scholar
  61. [61]
    Masry, E. (1996). Multivariate probabilty density estimation by wavelet methods: strong consistency and rates for stationary time series. Technical report, University of California, San Diego.Google Scholar
  62. [62]
    Meyer, Y. (1990).Ondelettes et Opéerateurs I: Ondelettes. Hermann, Paris.Google Scholar
  63. [63]
    Michelli, C. A. andRivlin, T. J. (1975). A survey of optimal recovery. In Michelli, C. A. and Rivlin, T. J. (eds.),Optimal estimation in Approximating theory, pp. 1–54, Plenum, New York.Google Scholar
  64. [64]
    Müller, H. G. (1985). Empirical bandwidth choice for nonparametric kernel regression by means of pilot estimators.Statist. Decisions 2, 193–206.Google Scholar
  65. [65]
    Nason, G. P. (1996). Wavelet regression using cross-validation.J. Roy. Statist. Soc., Ser B 58, 463–479.zbMATHMathSciNetGoogle Scholar
  66. [66]
    Nason, G. J. (1995). Choice of the threshold parameter in wavelet function estimation. In A. Antoniadis and G. Oppenheim (eds.),Wavelets and Statistics, pp. 261–280, Lecture Notes in Statistics, Springer-Verlag, New York.Google Scholar
  67. [67]
    Nason, G. J. andSilverman, B. W. (1994). The discrete wavelet transform inS.Journal of Computational and Graphical Statistics 3, 163–191.CrossRefGoogle Scholar
  68. [68]
    Neumann, M. H. (1994). Spectral density estimation via nonlinear wavelet methods for stationary non-Gaussian series. Technical report 99, Institute for applied and stochastic analysis, Berlin.Google Scholar
  69. [69]
    Neumann, M. H. andSpokoiny, V. G. (1993). On the efficiency of wavelet estimators under arbitrary error distributions.Discussion Paper No. 4, Hümboldt Universität zu Berlin.Google Scholar
  70. [70]
    Ogden, T. R. (1996).Essential wavelets for statistical applications and data analysis. Birkhäuser, Basel.Google Scholar
  71. [71]
    Oudshoorn, C. (1994). Wavelet-based nonparametric regression: optimal rate in the sup-norm. Technical report848, University Utrecht.Google Scholar
  72. [72]
    Penev, S. andDechevsky, L. (1997). On non-negative wavelet-based estimators. Technical report, University of New South Wales. To appear inJ. of Nonparam. Statistics.Google Scholar
  73. [73]
    Pinheiro, A. andVidakovic, B. (1995). Estimating the square root of a density via compactly supported wavelets. Technical reportDP 95-14, Duke University.Google Scholar
  74. [74]
    Potier, C. andVercken, C. (1994). Spline fitting Numerous Noisy Data with discontinuities. In Laurent et al. (eds.),Curves and Surfaces, pp. 477–480, Academic Press, New York.Google Scholar
  75. [75]
    Raimondo, M. (1996).Situations non ergodiques et utilisations de méthodes d’ondelettes. Ph. D. Thesis, University Paris 7.Google Scholar
  76. [76]
    Ramlau-Hansen, H. (1983). Smoothing counting processes by means of kernel functions.Ann. Statist.11, 453–466.zbMATHCrossRefMathSciNetGoogle Scholar
  77. [77]
    Saito, N. (1994). Simultaneous noise suppression and signal compression using a library of orthonormal bases and the minimum description length criterion. In Foufoula-Georgiou, E. and Kumar, P. (eds.),Wavelets in Geophysics, Academic Press, New York.Google Scholar
  78. [78]
    Stein, C. (1981). Estimation of the mean of a multivariate normal distribution.Ann. Statist.10, 1135–1151.CrossRefGoogle Scholar
  79. [79]
    Sweldens, W. (1996). The lifting scheme: A custom-design construction of biorthogonal wavelets.Applied and Comp. Harmonic Analysis 3, 186–200.zbMATHCrossRefMathSciNetGoogle Scholar
  80. [80]
    Sweldens, W. (1996). The lifting scheme: A construction of second generation wavelets. Technical report 1995:6, Industrial Mathematics Initiative, Department of Mathematics, University of South Carolina.Google Scholar
  81. [81]
    Triebel, H. (1992).Theory of function spaces II. Birkhäuser, Basel.Google Scholar
  82. [82]
    Tribouley, K. (1995). Practical estimation of multivariate density using wavelet methods.Statistica Neerlandica 49, 41–62.zbMATHCrossRefMathSciNetGoogle Scholar
  83. [83]
    Vannucci, M. andVidakovic, B. (1995). Preventing the Dirac disaster: wavelet based density estimation. Technical reportDP 95-24, Duke University.Google Scholar
  84. [84]
    Vidakovic, B. (1994). Nonlinear wavelet shrinkage with Bayes rules and Bayes factors. Technical reportDP 94-24, Duke University.Google Scholar
  85. [85]
    von Sachs, R. andSchneider, K. (1996). Wavelet smoothing of evolutionary spectra by nonlinear thresholding. Applied and Comp.Harmonic Analysis 3 (3), 268–282.zbMATHCrossRefMathSciNetGoogle Scholar
  86. [86]
    Wahba, G. (1990).Spline models for observational data. CBMS-NSF regional conferences series in applied mathematics. SIAM, Philadelphia.Google Scholar
  87. [87]
    Walter, G. G. (1992). Approximation of the Delta Function by Wavelets.J. Approx. Theory 71, 329–343.zbMATHCrossRefMathSciNetGoogle Scholar
  88. [88]
    Walter, G. O. (1994).Wavelets and Other Orthogonal Systems with Applications. CRC Press, Boca Raton, Florida.zbMATHGoogle Scholar
  89. [89]
    Wang, Y. (1995). Jump and Sharp Cusp Detection by wavelets.Biometrika 82, 385–397.zbMATHCrossRefMathSciNetGoogle Scholar
  90. [90]
    Wang, Y. (1996). Function estimation via wavelet shrinkage for long-memory data.Ann. Statist.24 (2), 466–484.zbMATHCrossRefMathSciNetGoogle Scholar
  91. [91]
    Weyrich, N. andWarhola, O. T. (1995). Denoising using wavelets and cross-validation. IN Singh, S. P. (ed.),Approximation Theory, wavelet and applications, NATO ASI series C, pp. 523–532.Google Scholar
  92. [92]
    Wickerhauser, M. V. (1994).Adapted Wavelet Analysis: From Theory to Software. AK Peters, Boston.zbMATHGoogle Scholar

Copyright information

© Società Italiana di Statistica 1997

Authors and Affiliations

  • A. Antoniadis
    • 1
  1. 1.Laboratoire IMAG-LMCUniversity Joseph FourierGrenoble Cedex 9France

Personalised recommendations