Empirical Bayes Methods for Smoothing Data and for Simultaneous Estimation of Many Parameters

A recent successful development is found in a series of innovative, new statistical methods for smoothing data that are based on the empirical Bayes method. This paper emphasizes their practical usefulness in medical sciences and their theoretically close relationship with the problem of simultaneou...

Full description

Saved in:
Bibliographic Details
Published inEnvironmental health perspectives Vol. 87; pp. 109 - 114
Main Authors Yanagimoto, Takemi, Kashiwagi, Nobuhisa
Format Journal Article
LanguageEnglish
Published United States National Institute of Environmental Health Sciences. National Institutes of Health. Department of Health, Education and Welfare 01.07.1990
Subjects
Application of Nonparametric Smoothing
Bayes Theorem
Clioquinol - adverse effects
Cohort Studies
Data models
Data smoothing
Estimation methods
Estimators
Female
Humans
Inference
Japan - epidemiology
Likelihood Functions
Linear regression
Male
Maximum likelihood estimation
Mortality
Neoplasms - mortality
Nervous System Diseases - chemically induced
Nervous System Diseases - epidemiology
Scientific method
Seasonal adjustment
Seasons
Stomach Neoplasms - mortality
Japan
Online AccessGet full text

Cover

More Information
Summary:A recent successful development is found in a series of innovative, new statistical methods for smoothing data that are based on the empirical Bayes method. This paper emphasizes their practical usefulness in medical sciences and their theoretically close relationship with the problem of simultaneous estimation of parameters, depending on strata. The paper also presents two examples of analyzing epidemiological data obtained in Japan using the smoothing methods to illustrate their favorable performance.
ISSN:0091-6765
1552-9924
DOI:10.1289/ehp.9087109