Large-Scale Multiple Testing for Spatially Clustered Data

• Activity Number: 629
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Mental Health Statistics Section
• Author(s): Hongyuan Cao*+ and Yunda Zhong and Wei Biao Wu
• Companies: the university of chicago and The University of Chicago and The University of Chicago
• Keywords: Multiple testing ; Statistical power ; Change point ; High dimension ; Nonstationary ; Clustered data

Abstract

We consider large scale multiple testing for clustered signals, where the signals exhibit spatial dependence structure. A change point boundary detection procedure is proposed to make use of the spatial information for hypothesis testing. It is assumed that the number of clusters for alternative hypotheses is finite and the clusters are well separated. We show that by exploiting the spatial structure, the precision of a multiple testing procedure can be improved substantially. Simulation studies evidence that the methods perform well with realistic sample sizes and demonstrate the improved detection ability compared with competing methods. The practical utility of the method is illustrated on data from a genomics study.