Making the Cut: Improved Ranking and Selection for Large-Scale Inference
Open Access
- 8 November 2015
- journal article
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 78 (4) , 781-804
Abstract
Summary: Identifying leading measurement units from a large collection is a common inference task in various domains of large-scale inference. Testing approaches, which measure evidence against a null hypothesis rather than effect magnitude, tend to overpopulate lists of leading units with those associated with low measurement error. By contrast, local maximum likelihood approaches tend to favour units with high measurement error. Available Bayesian and empirical Bayesian approaches rely on specialized loss functions that result in similar deficiencies. We describe and evaluate a generic empirical Bayesian ranking procedure that populates the list of top units in a way that maximizes the expected overlap between the true and reported top lists for all list sizes. The procedure relates unit-specific posterior upper tail probabilities with their empirical distribution to yield a ranking variable. It discounts high variance units less than popular non-maximum-likelihood methods and thus achieves improved operating characteristics in the models considered.Keywords
All Related Versions
Funding Information
- US National Institutes of Health (R21 HG006568, T32 GM074904, U54AI117924)
This publication has 27 references indexed in Scilit:
- EBSeq: an empirical Bayes hierarchical model for inference in RNA-seq experimentsBioinformatics, 2013
- Whole-Genome Regression and Prediction Methods Applied to Plant and Animal BreedingGenetics, 2013
- Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetesNature Genetics, 2012
- Testing significance relative to a fold-change threshold is a TREATBioinformatics, 2009
- Fundamental Differences in Cell Cycle Deregulation in Human Papillomavirus–Positive and Human Papillomavirus–Negative Head/Neck and Cervical CancersCancer Research, 2007
- Loss function based ranking in two-stage, hierarchical modelsBayesian Analysis, 2006
- Linear Models and Empirical Bayes Methods for Assessing Differential Expression in Microarray ExperimentsStatistical Applications in Genetics and Molecular Biology, 2004
- Loss functions for estimation of extrema with an application to disease mappingCanadian Journal of Statistics, 2003
- Bayes FactorsJournal of the American Statistical Association, 1995
- A Bayesian Approach to Ranking and Selection of Related Means with Alternatives to Analysis-of-Variance MethodologyJournal of the American Statistical Association, 1988