Abstract
In this article, we study empirical Bayes testing for uniform distributions based on randomly censored data. We construct an empirical Bayes test δ̃n, and study its associated asymptotic optimality. It is shown that under some conditions on the prior distribution π, δ̃n is asymptotically optimal and its associated regret converges to zero at a rate O(n−r/(r+1)), where r is a positive integer, depending on a condition on π and n is the number of past data available when the present testing problem is considered.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Földes, A., Rejtö, L., 1981. A LIL type result for the product limit estimators. Z. Wahrscheinlichkeitsthero. Verw Geb., 56, 75–86.
Liang, T., 2008. On empirical Bayes testing in a discrete life model with random censoring. Commun. Statist. Theory and Methods, 37 (15), 2369–2381.
Liang, T., Huang, W. T., 2006. Empirical Bayes testing with random censoring. Technical Report.
Liang, T., Huang, W. T., 2009. Empirical Bayes estimation for uniform distributions with random right censoring. To appear in Commun. Statist. Theory and Methods.
Robbins, H., 1956. An empirical Bayes approach to statistics. In Proc. Third Berkeley Symp. Math. Statist. and Probab., University of California Press: Berkeley, CA, Vol. 1, 157–163.
Susarla, V., van Ryzin, J., 1978. Empirical Bayes estimation of a distribution (survival) function from right censored observations. Ann. Statist., 6, 740–754.
Susarla, V., van Ryzin, J., 1986. Empirical Bayes procedures with censored data. In Adaptive Statistical Procedures and Related Topics, J. van Ryzin (editor), IMS Lecture Notes-Monograph Series, 8, 219–234.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Liang, T. Empirical Bayes Testing for Uniform Distributions with Random Censoring. J Stat Theory Pract 2, 633–649 (2008). https://doi.org/10.1080/15598608.2008.10411899
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1080/15598608.2008.10411899