Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A novel weighted likelihood estimation with empirical Bayes flavor

Hossain, Mobarak ; Kozubowski, Tomasz and Podgorski, Krzysztof LU (2015) In Working Papers in Statistics
Abstract
We propose a novel approach to estimation, where each individual observation in a random sample is used to derive an estimator of an unknown parameter using the maximum likelihood principle. These individual estimators are then combined as a weighted average to produce the final estimator. The weights are chosen to be proportional to the likelihood function evaluated at the estimators based on each observation. The method can be related to a Bayesian approach, where the prior distribution is data driven. In case of estimating a location parameter of a unimodal density, the prior distribution is the empirical distribution of the sample, and converges to the true distribution that generated the data as the sample size increases.

... (More)
We propose a novel approach to estimation, where each individual observation in a random sample is used to derive an estimator of an unknown parameter using the maximum likelihood principle. These individual estimators are then combined as a weighted average to produce the final estimator. The weights are chosen to be proportional to the likelihood function evaluated at the estimators based on each observation. The method can be related to a Bayesian approach, where the prior distribution is data driven. In case of estimating a location parameter of a unimodal density, the prior distribution is the empirical distribution of the sample, and converges to the true distribution that generated the data as the sample size increases.



We provide several examples illustrating the new method, argue for its consistency, and conduct simulation studies to assess the performance of the estimators. It turns out that this straightforward methodology produces consistent estimators, which seem to be comparable with those obtained by the maximum likelihood method. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Working paper/Preprint
publication status
published
subject
keywords
Consistency, data-dependent prior, empirical Bayes, exponentiated distribution, location parameter, maximum likelihood estimator, super-efficiency, unbounded likelihood
in
Working Papers in Statistics
issue
6
pages
28 pages
publisher
Department of Statistics, Lund university
language
English
LU publication?
yes
id
ccf31357-b6ca-4608-9fd6-6104d952c20e (old id 8052689)
alternative location
http://journals.lub.lu.se/index.php/stat/article/view/15037
date added to LUP
2016-04-04 09:58:09
date last changed
2018-11-21 20:55:56
@misc{ccf31357-b6ca-4608-9fd6-6104d952c20e,
  abstract     = {{We propose a novel approach to estimation, where each individual observation in a random sample is used to derive an estimator of an unknown parameter using the maximum likelihood principle. These individual estimators are then combined as a weighted average to produce the final estimator. The weights are chosen to be proportional to the likelihood function evaluated at the estimators based on each observation. The method can be related to a Bayesian approach, where the prior distribution is data driven. In case of estimating a location parameter of a unimodal density, the prior distribution is the empirical distribution of the sample, and converges to the true distribution that generated the data as the sample size increases.<br/><br>
<br/><br>
We provide several examples illustrating the new method, argue for its consistency, and conduct simulation studies to assess the performance of the estimators. It turns out that this straightforward methodology produces consistent estimators, which seem to be comparable with those obtained by the maximum likelihood method.}},
  author       = {{Hossain, Mobarak and Kozubowski, Tomasz and Podgorski, Krzysztof}},
  keywords     = {{Consistency; data-dependent prior; empirical Bayes; exponentiated distribution; location parameter; maximum likelihood estimator; super-efficiency; unbounded likelihood}},
  language     = {{eng}},
  note         = {{Working Paper}},
  number       = {{6}},
  publisher    = {{Department of Statistics, Lund university}},
  series       = {{Working Papers in Statistics}},
  title        = {{A novel weighted likelihood estimation with empirical Bayes flavor}},
  url          = {{https://lup.lub.lu.se/search/files/5429067/8054204.pdf}},
  year         = {{2015}},
}