Density estimation by wavelet thresholding
Donoho, David L. ; Johnstone, Iain M. ; Kerkyacharian, Gérard ; Picard, Dominique
Ann. Statist., Tome 24 (1996) no. 6, p. 508-539 / Harvested from Project Euclid
Density estimation is a commonly used test case for nonparametric estimation methods. We explore the asymptotic properties of estimators based on thresholding of empirical wavelet coefficients. Minimax rates of convergence are studied over a large range of Besov function classes $B_{\sigma pq}$ and for a range of global $L'_p$ error measures, $1 \leq p' < \infty$. A single wavelet threshold estimator is asymptotically minimax within logarithmic terms simultaneously over a range of spaces and error measures. In particular, when $p' > p$, some form of nonlinearity is essential, since the minimax linear estimators are suboptimal by polynomial powers of n. A second approach, using an approximation of a Gaussian white-noise model in a Mallows metric, is used to attain exactly optimal rates of convergence for quadratic error $(p' = 2)$.
Publié le : 1996-04-14
Classification:  Minimax estimation,  adaptive estimation,  density estimation,  spatial adaptation,  wavelet orthonormal bases,  Besov spaces,  62G07,  62G20
@article{1032894451,
     author = {Donoho, David L. and Johnstone, Iain M. and Kerkyacharian, G\'erard and Picard, Dominique},
     title = {Density estimation by wavelet thresholding},
     journal = {Ann. Statist.},
     volume = {24},
     number = {6},
     year = {1996},
     pages = { 508-539},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1032894451}
}
Donoho, David L.; Johnstone, Iain M.; Kerkyacharian, Gérard; Picard, Dominique. Density estimation by wavelet thresholding. Ann. Statist., Tome 24 (1996) no. 6, pp.  508-539. http://gdmltest.u-ga.fr/item/1032894451/