On optimal convergence rate of empirical Bayes tests
This paper deals with the optimal convergence rate of empirical Bayes tests. A sharp lower bound on the minimax regret of empirical Bayes tests is established. When the result is applied to the normal distribution model, a lower convergence rate n-1 is obtained. We also construct an empirical Bayes test [delta]n* whose corresponding regret achieves the rate n-1, provided that [psi]G, the characteristic function of the prior distribution G, is such that [psi]G(t)=0 for t[greater-or-equal, slanted]b-1 for some b>0. Therefore, for the empirical Bayes testing for a normal mean problem, the optimal convergence rate is n-1.
Volume (Year): 68 (2004)
Issue (Month): 2 (June)
|Contact details of provider:|| Web page: http://www.elsevier.com/wps/find/journaldescription.cws_home/622892/description#description|
|Order Information:|| Postal: http://www.elsevier.com/wps/find/supportfaq.cws_home/regional|
When requesting a correction, please mention this item's handle: RePEc:eee:stapro:v:68:y:2004:i:2:p:189-198. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Zhang, Lei)
If references are entirely missing, you can add them using this form.