报告题目：Bias and variance reduction in estimating theproportion of true null hypotheses

报告人：童铁军教授（香港浸会大学）

报告时间：2015年9月17日下午3:30-4:30

报告地点：知新楼B-1238

摘要：

When testing a large number of hypotheses,estimating the proportion of true nulls, denoted by pi0, becomes increasinglyimportant. This quantity has many applications in practice. For instance, areliable estimate of pi0 can eliminate the conservative bias of the Benjamini–Hochberg procedure on controlling the falsediscovery rate. It is known that most methods in the literature for estimatingpi0 are conservative. Recently, some attempts have been paid to reduce suchestimation bias. Nevertheless, they are either over bias corrected or sufferingfrom an unacceptably large estimation variance. In this paper, we propose a newmethod for estimating pi0 that aims to reduce the bias and variance of theestimation simultaneously. To achieve this, we first utilize the probabilitydensity functions of false-null p-values and then propose a novel algorithm toestimate the quantity of pi0. The statistical behavior of the proposedestimator is also investigated. Finally, we carry out extensive simulationstudies and several real data analysis to evaluate the performance of theproposed estimator. Both simulated and real data demonstrate that the proposedmethod may improve the existing literature significantly.

热烈欢迎各位老师和同学参加！