In the framework of a wavelet analysis, we study the statistical meaning of many classes of procedures. More precisely, we aim at investigating the maximal spaces (maxisets) where these procedures attain a given rate of convergence. The maxiset approach allows to bring theoretical explanations on some phenomena observed in the practical setting which are not explained by the minimax approach. Indeed, we show that data-driven thresholding rules outperform non random thresholding rules. Then, we prove that procedures which consist in thresholding coefficients by groups, as tree rules (close to Lepski's rule) or block thresholding rules, are often better in the maxiset sense than procedures which consist in thresholding coefficients individual...
International audienceWe focus on the performances of tree-structured wavelet estimators belonging t...
In this paper, we address the situation where we cannot differentiate wavelet-based threshold proced...
International audienceIn this paper we compute the maxisets of some denoising methods (estimators) f...
We study the performance of a large collection of block thresholding wavelet estimators, namely the ...
International audienceWe study the maxiset performance of a large collection of block thresholding w...
International audienceWe study the maxiset performance of a large collection of block thresholding w...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
L. Birgé, L. Elie, Y. Golubev (rapporteur), G. Kerkyacharian, O. Lepski, D. Picard, A. Tsybakov, S. ...
International audienceIn this paper, we address the situation where we cannot differentiate wavelet-...
International audienceIn this paper, we address the situation where we cannot differentiate wavelet-...
L. Birgé, L. Elie, Y. Golubev (rapporteur), G. Kerkyacharian, O. Lepski, D. Picard, A. Tsybakov, S. ...
In this paper our aim is to provide tools for easily calculating the maxisets of several procedures....
International audienceWe focus on the performances of tree-structured wavelet estimators belonging t...
In this paper, we address the situation where we cannot differentiate wavelet-based threshold proced...
International audienceIn this paper we compute the maxisets of some denoising methods (estimators) f...
We study the performance of a large collection of block thresholding wavelet estimators, namely the ...
International audienceWe study the maxiset performance of a large collection of block thresholding w...
International audienceWe study the maxiset performance of a large collection of block thresholding w...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
International audienceWe address the statistical issue of determining the maximal spaces (maxisets) ...
L. Birgé, L. Elie, Y. Golubev (rapporteur), G. Kerkyacharian, O. Lepski, D. Picard, A. Tsybakov, S. ...
International audienceIn this paper, we address the situation where we cannot differentiate wavelet-...
International audienceIn this paper, we address the situation where we cannot differentiate wavelet-...
L. Birgé, L. Elie, Y. Golubev (rapporteur), G. Kerkyacharian, O. Lepski, D. Picard, A. Tsybakov, S. ...
In this paper our aim is to provide tools for easily calculating the maxisets of several procedures....
International audienceWe focus on the performances of tree-structured wavelet estimators belonging t...
In this paper, we address the situation where we cannot differentiate wavelet-based threshold proced...
International audienceIn this paper we compute the maxisets of some denoising methods (estimators) f...