Abstract
Statistical methods involving high-dimensional testing, i.e., a large number of simultaneous tests, have become important in recent days. Applications include microarrays, fMRI images and signal processing. Information that can be obtained by treating them as connected tests leads to the concept of discovery rates as well as to the Bayesian approach to hypothesis tests. It gives us great pleasure to honour Herbert Robbins who introduced the Empirical Bayes technique in statistical inference, which connects the frequentist and the Bayesian approaches in this problem.
Similar content being viewed by others
Suggested Reading
B Efron, R Tibshirani, J D Storey and V Tusher, Empirical Bayes analysis of a microarray experiment, J. Amer. Statist. Assoc., Vol.96, pp.1151–1160, 2001.
Y Benjamini and Y Hochberg, Controlling the false discovery rate: A practical and powerful approach to multiple testing, J. Roy. Statist. Soc. Ser. B, Vol.57, pp.289–300, 1995.
S Ghosal and A Roy, Bayesian nonparametric approach to multiple testing, Perspectives in Mathematical Sciences — I, pp.139–164, 2009.
M Delampady and T Krishnan, Bayesian statistics, Resonance, Vol.7, No.4, pp.27–38, 2002.
J D Storey, Adirect approach to false discovery rates, J. Roy. Statist. Soc. Ser. B, Vol.64, pp.479–498, 2002.
Author information
Authors and Affiliations
Corresponding author
Additional information
Soumen Dey is a research scholar at ISI, Bangalore.
Mohan Delampady is a Professor at the Indian Statistical Institute, Bangalore.
Rights and permissions
About this article
Cite this article
Dey, S., Delampady, M. False discovery rates and multiple testing. Reson 18, 1095–1109 (2013). https://doi.org/10.1007/s12045-013-0137-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12045-013-0137-9