The Annals of Statistics

Wavelet-based estimation with multiple sampling rates

Peter Hall and Spiridon Penev

Full-text: Open access

Abstract

We suggest an adaptive sampling rule for obtaining information from noisy signals using wavelet methods. The technique involves increasing the sampling rate when relatively high-frequency terms are incorporated into the wavelet estimator, and decreasing it when, again using thresholded terms as an empirical guide, signal complexity is judged to have decreased. Through sampling in this way the algorithm is able to accurately recover relatively complex signals without increasing the long-run average expense of sampling. It achieves this level of performance by exploiting the opportunities for near-real time sampling that are available if one uses a relatively high primary resolution level when constructing the basic wavelet estimator. In the practical problems that motivate the work, where signal to noise ratio is particularly high and the long-run average sampling rate may be several hundred thousand operations per second, high primary resolution levels are quite feasible.

Article information

Source
Ann. Statist., Volume 32, Number 5 (2004), 1933-1956.

Dates
First available in Project Euclid: 27 October 2004

Permanent link to this document
https://projecteuclid.org/euclid.aos/1098883777

Digital Object Identifier
doi:10.1214/009053604000000751

Mathematical Reviews number (MathSciNet)
MR2102498

Zentralblatt MATH identifier
1056.62043

Subjects
Primary: 62G08: Nonparametric regression
Secondary: 62P30: Applications in engineering and industry

Keywords
Adaptive estimation bandwidth digital recording high frequency nonparametric regression online analysis primary resolution level sampling rule signal analysis threshold sequential analysis

Citation

Hall, Peter; Penev, Spiridon. Wavelet-based estimation with multiple sampling rates. Ann. Statist. 32 (2004), no. 5, 1933--1956. doi:10.1214/009053604000000751. https://projecteuclid.org/euclid.aos/1098883777


Export citation

References

  • Aldroubi, A. and Gröchenig, K. (2001). Nonuniform sampling and reconstruction in shift-invariant spaces. SIAM Rev. 43 585--620.
  • Anon (2002). Mathematics: A new wave. The Economist January 19 70.
  • Antoniadis, A. and Fan, J. Q. (2001). Regularization of wavelet approximations (with discussion). J. Amer. Statist. Assoc. 96 939--967.
  • Cai, T. T. and Brown, L. D. (1998). Wavelet shrinkage for nonequispaced samples. Ann. Statist. 26 1783--1799.
  • Cai, T. T. and Brown, L. D. (1999). Wavelet estimation for samples with random uniform design. Statist. Probab. Lett. 42 313--321.
  • Chen, W., Itoh, S. and Shiki, J. (1998). Irregular sampling theorems for wavelet subspaces. IEEE Trans. Inform. Theory 44 1131--1142.
  • Delouille, V., Franke, J. and von Sachs, R. (2001). Nonparametric stochastic regression with design-adapted wavelets. Sankhyā Ser. A 63 328--366.
  • Delouille, V., Simoens, J. and von Sachs, R. (2001). Smooth design-adapted wavelets for nonparametric stochastic regression. Discussion Paper No. 0117, l'Institut de Statistique, Univ. Catholique de Louvain, Belgium.
  • Delouille, V. and von Sachs, R. (2002). Properties of design-adapted wavelet transforms of nonlinear autoregression models. Discussion Paper No. 0225, l'Institut de Statistique, Univ. Catholique de Louvain, Belgium.
  • Donoho, D. L. and Johnstone, I. M. (1994). Ideal spatial adaptation by wavelet shrinkage. Biometrika 81 425--455.
  • Donoho, D. L. and Johnstone, I. M. (1995). Adapting to unknown smoothness via wavelet shrinkage. J. Amer. Statist. Assoc. 90 1200--1224.
  • Donoho, D. L., Johnstone, I. M., Kerkyacharian, G. and Picard, D. (1995). Wavelet shrinkage: Asymptopia? (With discussion.) J. Roy. Statist. Soc. Ser. B 57 301--369.
  • Hall, P. and Penev, S. (2002). Wavelet-based estimation with multiple sampling rates. Available at www.maths.unsw.edu.au/~spiro/publicat.html.
  • Hall, P. and Patil, P. (1995). Formulae for mean integrated squared error of nonlinear wavelet-based density estimators. Ann. Statist. 23 905--928.
  • Hall, P., Park, B. U. and Turlach, B. A. (1998). A note on design transformation and binning in nonparametric curve estimation. Biometrika 85 469--476.
  • Hall, P. and Turlach, B. A. (1997). Interpolation methods for nonlinear wavelet regression with irregularly spaced design. Ann. Statist. 25 1912--1925.
  • Herzberg, A. M. and Traves, W. N. (1994). An optimal experimental design for the Haar regression model. Canad. J. Statist. 22 357--364.
  • Kovac, A. and Silverman, B. W. (2000). Extending the scope of wavelet regression methods by coefficient-dependent thresholding. J. Amer. Statist. Assoc. 95 172--183.
  • Oyet, A. J. and Wiens, D. P. (2000). Robust designs for wavelet approximations of regression models. J. Nonparametr. Statist. 12 837--859.
  • Pensky, M. and Vidakovic, B. (2001). On non-equally spaced wavelet regression. Ann. Inst. Statist. Math. 53 681--690.
  • Sardy, S., Percival, D. B., Bruce, A. G., Gao, H. Y. and Stuetzle, W. (1999). Wavelet shrinkage for unequally spaced data. Statist. Comput. 9 65--75.
  • Zhang, S. L. and Zheng, Z. G. (1999). Nonlinear wavelet estimation of regression function with random design. Sci. China Ser. A 42 825--833.