Premium
An improved empirical Bayes test for positive exponential families
Author(s) -
Liang Tachen
Publication year - 2002
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/1467-9574.t01-1-00070
Subject(s) - bayes' theorem , mathematics , regret , rate of convergence , exponential function , statistics , convergence (economics) , zero (linguistics) , exponential family , empirical research , econometrics , bayesian probability , computer science , mathematical analysis , economics , computer network , channel (broadcasting) , linguistics , philosophy , economic growth
We exhibit an empirical Bayes test δ * n for a decision problem using a linear error loss in a class of positive exponential families. This empirical Bayes test δ * n possesses the asymptotic optimality, and its associated regret converges to zero with rate n −1 (ln n ) 6 This rate of convergence improves the previous results in the literature in the sense that a faster rate of convergence is achieved under much weaker conditions. Examples are presented to illustrate the performance of the empirical Bayes test δ * n