Iterative estimates for a smoothing parameter
A recently developed penalized non-parametric maximum likehood estimate (NPMLE), of a non-increasing density, overcomes the so-called spiking problem in the well-known NPMLE. In this paper, some iterative procedures for choosing the optimal smoothing parameter in the penalized NPMLE are considered. The iterative estimates are shown to converge and improve on adaptive estimates. Comparisons with Jackknife estimates are made. Remarks about bootstrap and kernel estimates are given.
Year of publication: |
1995
|
---|---|
Authors: | Sun, Jiayang |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 24.1995, 4, p. 347-356
|
Publisher: |
Elsevier |
Keywords: | Penalized estimates Non-parametric maximum likelihood Iterative and adaptive estimates |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Sun, Jiayang, (2009)
-
Testing homogeneity in a mixture distribution via the L2 distance between competing models
Charnigo, Richard, (2004)
-
Charnigo, Richard, (2004)
- More ...