On the performance of isotropic and hyperbolic wavelet estimators
AbstractIn this paper, we differentiate between isotropic and hyperbolic wavelet bases in the context of multivariate nonparametric function estimation. The study of the latter leads to new phenomena and non trivial extensions of univariate studies. In this context, we first exhibit the limitations of isotropic wavelet estimators by proving that no isotropic estimator is able to guarantee the reconstruction of a function with anisotropy in an optimal or near optimal way. Second, we show that hyperbolic wavelet estimators are well suited to reconstruct anisotropic functions. In particular, for each considered estimator we focus on the rates at which it can reconstruct functions from anisotropic Besov spaces. We then compute the estimator's maxiset, this is the largest functional spaces over which its risk converges at these rates. Our results furnish novel arguments to understand the primordial role of sparsity and thresholding in multivariate contexts, notably by showing the exposure of linear methods to the curse of dimensionality. Moreover, we propose a block thresholding hyperbolic estimator and show its ability to estimate anisotropic functions at the optimal minimax rate and related, the remaining pertinence of information pooling in high dimensional settings.
Download InfoIf you experience problems downloading a file, check if you have the proper application to view it first. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please be patient as the files may be large.
Bibliographic InfoPaper provided by Katholieke Universiteit Leuven in its series Open Access publications from Katholieke Universiteit Leuven with number urn:hdl:123456789/377183.
Length: 31 pages
Date of creation: Jan 2013
Date of revision:
Contact details of provider:
Web page: http://www.kuleuven.be
Anisotropy; Besov spaces; Minimax and maxiset approaches; Linear and non linear methods; Thresholding rules; Multidimensional wavelet basis;
This paper has been announced in the following NEP Reports:
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Gérard Kerkyacharian & Dominique Picard & Lucien Birgé & Peter Hall & Oleg Lepski & Enno Mammen & Alexandre Tsybakov & G. Kerkyacharian & D. Picard, 2000. "Thresholding algorithms, maxisets and well-concentrated bases," TEST: An Official Journal of the Spanish Society of Statistics and Operations Research, Springer, vol. 9(2), pages 283-344, December.
- Autin, F. & Claeskens, Gerda & Freyermuth, Jean-Marc, 2012. "Hyperbolic wavelet thresholding methods and the curse of dimensionality through the maxiset approach," Open Access publications from Katholieke Universiteit Leuven urn:hdl:123456789/356464, Katholieke Universiteit Leuven.
- Genest, Christian & Masiello, Esterina & Tribouley, Karine, 2009. "Estimating copula densities through wavelets," Insurance: Mathematics and Economics, Elsevier, vol. 44(2), pages 170-181, April.
- Florent Autin, 2008. "Maxisets for μ-thresholding rules," TEST: An Official Journal of the Spanish Society of Statistics and Operations Research, Springer, vol. 17(2), pages 332-349, August.
- Arnak Dalalyan & Yuri Ingster & Alexandre B. Tsybakov, 2012. "Statistical Inference in Compound Functional Models," Working Papers 2012-20, Centre de Recherche en Economie et Statistique.
- Autin, F. & Le Pennec, E. & Tribouley, K., 2010. "Thresholding methods to estimate copula density," Journal of Multivariate Analysis, Elsevier, vol. 101(1), pages 200-222, January.
- Anestis Antoniadis & Jeremie Bigot & Theofanis Sapatinas, . "Wavelet Estimators in Nonparametric Regression: A Comparative Simulation Study," Journal of Statistical Software, American Statistical Association, vol. 6(i06).
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Carl Demeyere).
If references are entirely missing, you can add them using this form.