## Wavelet Thresholding via a Bayesian Approach (1996)

### Cached

### Download Links

Venue: | J. R. STATIST. SOC. B |

Citations: | 201 - 27 self |

### BibTeX

@ARTICLE{Abramovich96waveletthresholding,

author = {F. Abramovich and T. Sapatinas and B. W. Silverman},

title = {Wavelet Thresholding via a Bayesian Approach},

journal = {J. R. STATIST. SOC. B},

year = {1996},

volume = {60},

pages = {725--749}

}

### Years of Citing Articles

### OpenURL

### Abstract

We discuss a Bayesian formalism which gives rise to a type of wavelet threshold estimation in non-parametric regression. A prior distribution is imposed on the wavelet coefficients of the unknown response function, designed to capture the sparseness of wavelet expansion common to most applications. For the prior specified, the posterior median yields a thresholding procedure. Our prior model for the underlying function can be adjusted to give functions falling in any specific Besov space. We establish a relation between the hyperparameters of the prior model and the parameters of those Besov spaces within which realizations from the prior will fall. Such a relation gives insight into the meaning of the Besov space parameters. Moreover, the established relation makes it possible in principle to incorporate prior knowledge about the function's regularity properties into the prior model for its wavelet coefficients. However, prior knowledge about a function's regularity properties might b...

### Citations

2385 | A theory for multiresolution signal decomposition: The wavelet representation - Mallat - 1989 |

854 | Ideal spatial adaptation by wavelet shrinkage - Donoho, Johnstone - 1994 |

682 | Adapting to unknown smoothness via wavelet shrinkage - Donoho, Johnstone - 1995 |

412 | Theory of Function Spaces - Triebel - 1983 |

325 | Wavelet-based statistical signal processing using hidden Markov models - Crouse, Nowak, et al. - 1998 |

308 |
Wavelets and operators
- Meyer
- 1992
(Show Context)
Citation Context ... space of functions of bounded variation, sandwiched between B 1 1;1 and B 1 1;1 . The latter functions are of statistical interest because they allow for better models of spatial inhomogeneity (e.g. =-=Meyer, 1992-=-; Donoho & Johnstone, 1995). The Besov norm for the function g is related to a sequence space norm on the wavelet coefficients of the function. As noted in Section 2.1, confining attention to the reso... |

293 |
Estimation of the mean of a multivariate normal distribution
- Stein
- 1981
(Show Context)
Citation Context ...d level. Several data-driven thresholding rules have been developed recently. Donoho & Johnstone (1995) proposed a SureShrink thresholding which is based on minimizing Stein's unbiased risk estimate (=-=Stein, 1981-=-) and will usually yield smaller thresholds than the VisuShrink method. They have shown that SureShrink is also asymptotically near-minimax and the computational effort of the overall procedure is O(n... |

246 | Minimax estimation via wavelet shrinkage - Donoho, Johnstone - 1998 |

184 | Some aspects of the spline smoothing approach to non-parametric regression curve fitting (with discussion - Silverman - 1985 |

182 | Wavelet threshold estimators for data with correlated noise - Johnstone, Silverman - 1997 |

173 | Adaptive Bayesian wavelet shrinkage - Chipman, Kolaczyk, et al. - 1997 |

151 | Wavelet shrinkage: Asymptopia? (with discussion - Donoho, Johnstone, et al. - 1995 |

147 | Compression of wavelet decompositions - DeVore, Jawerth, et al. - 1992 |

135 | G: A correspondence between bayesian estimation on stochastic processes and smoothing by splines - GS, Wahba - 1970 |

118 | Multiple shrinkage and subset selection in wavelets - Clyde, Parmigiani, et al. - 1998 |

117 | W.: An overview of wavelet based multiresolution analyses - Jawerth, Sweldens |

115 | Nonlinear wavelet shrinkage with Bayes rules and Bayes factors
- Vidakovic
- 1998
(Show Context)
Citation Context ...pplications. Then, the function is estimated by applying some Bayes rules on the resulting posterior distribution of the wavelet coefficients. The traditional Bayes rule considered in the literature (=-=Vidakovic, 1994-=-; Clyde et al., 1995; Chipman et al., 1997) corresponds to an L 2 -loss (the posterior mean) based on the wavelet coefficients. However, such a rule is not a thresholding rule but a shrinkage. In orde... |

97 | P.: Multiresolution analysis, wavelets, and fast algorithms on an interval - Cohen, Daubechies, et al. - 1993 |

96 | Interpolation of Besov spaces - DeVore, Popov |

81 | The discrete wavelet transform in S - Nason, Silverman - 1994 |

77 | Wavelet shrinkage using cross-validation
- Nason
- 1996
(Show Context)
Citation Context ... wavelet-based estimators for reconstructing the original functions: VisuShrink (Donoho & Johnstone, 1994), GlobalSures(a modified version of the SureShrink of Donoho & Johnstone, 1995, considered in =-=Nason, 1996-=-), Cross-validation (Nason, 1995; 1996) and False Discovery Rate (Abramovich & Benjamini, 1995; 1996). Daubechies's least asymmetric wavelet of order 8 (defined through a set of sixteen non-zero coeff... |

69 | Wavelet Transforms versus Fourier Transforms - Strang - 1993 |

49 | Adaptive thresholding of wavelet coefficients - Abramovich, Benjamini - 1996 |

35 | Minimax Bayes, asymptotic minimax and sparse wavelet priors
- Johnstone
- 1994
(Show Context)
Citation Context ...The choice of thresholding rule, therefore, becomes a crucial step in the estimation procedure. Several approaches to thresholding have been introduced in the literature: a minimax approach (Donoho & =-=Johnstone, 1994-=-, 1995), multiple hypothesis testing (Abramovich & Benjamini, 1995, 1996; Ogden & Parzen, 1996a, 1996b), cross-validation (Nason, 1995, 1996; Weyrich & Warhola, 1995). The idea of thresholding has als... |

33 | Function estimation via Wavelet Shrinkage for Long-Memory Data - Wang - 1996 |

30 | Thresholding of wavelet coefficients as multiple hypotheses testing procedure - Abramovich, Benjamini - 1995 |

23 | Choice of the threshold parameter in wavelet function estimation
- Nason
- 1995
(Show Context)
Citation Context ...en introduced in the literature: a minimax approach (Donoho & Johnstone, 1994, 1995), multiple hypothesis testing (Abramovich & Benjamini, 1995, 1996; Ogden & Parzen, 1996a, 1996b), cross-validation (=-=Nason, 1995, 199-=-6; Weyrich & Warhola, 1995). The idea of thresholding has also been studied in the context of correlated errors " i ; see, for example, Wang (1996), Johnstone & Silverman (1997). In this paper we... |

18 | Data dependent wavelet thresholding in non-parametric regression with change-points applications - Odgen, Parzen - 1996 |

17 | Change-point approach to data analytic wavelet thresholding - Ogden, Parzen - 1996 |

13 |
The WaveThresh package: wavelet transform and thresholding software for S-Plus and R. Available from Statlib
- Nason
- 1993
(Show Context)
Citation Context ...and inverse DWT are performed by Mallat's (1989) fast algorithm that requires only O(n) operations and is available in several standard implementations, for example, in the S-plus package WaveThresh (=-=Nason, 1993-=-; Nason & Silverman, 1994). The WaveThresh package implements a periodized form of the DWT that produces (n \Gamma 1) sample discrete wavelet coefficients d jk , j = 0; : : : ; J \Gamma 1; k = 0; : : ... |

11 |
Probability
- Karr
- 1992
(Show Context)
Citation Context ...if and only if: 1 X j=0 2 js 0 q q j jj j jj q p = 1 X j=0 2 j ffiq j flq=2 jj j jj q p ! 1 almost surely. (21) It can be shown from the monotone convergence theorem and the three series theorem (see =-=Karr, 1993-=-, Theorems 4.10 and 7.5 respectively) that, if Z n are independent and identically distributed nonnegative random variables with strictly positive finite mean, and a n are nonnegative constants, then ... |

7 | Thresholding of wavelet coecients as multiple hypotheses testing procedure - Abramovich, Benjamini - 1995 |

6 | Adaptive thresholding of wavelet coe cients - Abramovich, Benjamini - 1996 |

5 | Multiparameter Univariate Bayesian Analysis - Stewart - 1979 |

1 | Data dependentwavelet thresholding in nonparametric regression with change-point applications - Ogden - 1996 |