Institution: Tel Aviv University
Street: Ramat Aviv
City: Tal Aviv
Authors: Daniel Yekutieli and Yoav Benjamini
Title: False Discovery Rate confidence Intervals
Abstract: In our presentation we will introduce a new method for constructing confidence intervals in multiple parameter estimation problems. The suggested intervals are not built for all parameters, but only for selected parameters, and are larger than the standard marginal 1-q confidence intervals but smaller than the multivariate intervals offering simultaneous 1-q covering probability. Our presentation will focus on the theoretical properties of the FDR confidence intervals and their relation other types of confidence intervals.