Brown University -
Division of Biology and Medicine,
Center for Statistical Sciences SEminar
Candidate for Assistant Professor (tenure track) in the Public Health Program/Department of Community Health | |
Abstract: The false discovery rate (FDR) procedure introduced by Benjamini and Hochberg in 1995 has become a mainstream method for large scale simultaneous inference. The control of the procedure is contingent upon the assumption that null p-values are uniform. In this talk, we discuss power and control of the procedure when that assumption is violated. Our key example is drawn from a SNPs association analysis.
This is joint work with Kathryn Roeder and Larry Wasserman.
<--- 2005 Index