Wavelet density estimation by approximation of log-densities
Probability density estimation is considered when log-density function belongs to the Besov function class Bspq. It is shown that n-2s/(2s+1) is a lower rate of convergence in Kullback-Leibler distance. Density functions are estimated by the maximum likelihood method in sequences of regular exponential families based on wavelet basis functions.
Year of publication: |
1996
|
---|---|
Authors: | Koo, Ja-Yong ; Kim, Woo-Chul |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 26.1996, 3, p. 271-278
|
Publisher: |
Elsevier |
Keywords: | Log-density estimation Exponential family Wavelet basis Besov spaces Rate |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Estimation of a regression function with a sharp change point using boundary wavelets
Park, Cheol-Woo, (2004)
-
The intrinsic distribution and selection bias of long-period cometary orbits
Jupp, Peter E., (2003)
-
Sharp adaptation for spherical inverse problems with applications to medical imaging
Koo, Ja-Yong, (2008)
- More ...