Weighted False Discovery Rate Control in Large- Scale Multiple Testing

By Pallavi Basu, Cai, T Tony, Das, Kiranmoy, Sun, Wenguang
Journal of the American Statistical Association, Theory and Methods [Ranked A* in Mathematical Sciences] | June 2018

DOI

http://www.tandfonline.com/doi/full/10.1080/01621459.2017.1336443

Access full text

Citation

Basu, Pallavi., Cai, T Tony., Das, Kiranmoy., Sun, Wenguang. Weighted False Discovery Rate Control in Large- Scale Multiple Testing Journal of the American Statistical Association, Theory and Methods [Ranked A* in Mathematical Sciences] http://www.tandfonline.com/doi/full/10.1080/01621459.2017.1336443.

Journal of the American Statistical Association, Theory and Methods [Ranked A* in Mathematical Sciences], 2018

Abstract

The use of weights provides an effective strategy to incorporate prior domain knowledge in large-scale inference. This article studies weighted multiple testing in a decision-theoretical framework. We develop oracle and data-driven procedures that aim to maximize the expected number of true positives subject to a constraint on the weighted false discovery rate. The asymptotic validity and optimality of the proposed methods are established. The results demonstrate that incorporating informative domain knowledge enhances the interpretability of results and precision of inference. Simulation studies show that the proposed method controls the error rate at the nominal level, and the gain in power over existing methods is substantial in many settings. An application to a genome-wide association study is discussed. Supplementary materials for this article are available online.

Pallavi Basu is an Assistant Professor of Operations Management at the Indian School of Business (ISB), where she teaches concepts and approaches in Statistics. Her research interests include the application of statistics in finance, marketing, and other disciplines; high-dimensional statistical inference; large-scale multiple testing; and topics on causal inference.

Professor Basu is a member of the American Statistical Association, the Institute of Mathematical Statistics, and the International Indian Statistical Association. She received her PhD in Business Administration and Statistics from the USC Marshall School of Business and was a postdoctoral fellow at Tel Aviv University. She completed her undergraduate and postgraduate studies in Statistics (specialising in mathematical statistics and probability) from the Indian Statistical Institute (ISI), Kolkata.

Her current research (2023-2026) is partly funded by the Mathematical Research Impact Centric Support (MATRICS) from the Science and Engineering Research Board (SERB), Government of India.

More publications by author

Pallavi Basu