Post‐selection point and interval estimation of signal sizes in Gaussian samples
Authors:
- 13 April 2017
- journal article
- research article
- Published by Wiley in Canadian Journal of Statistics
- Vol. 45 (2) , 128-148
Abstract
We tackle the problem of the estimation of a vector of underlying means (signal sizes) from a single vector‐valued observationy. Often one is interested in estimating only a subvector of signals corresponding to a set of selected, “interesting” sample elements. These “interesting” sample elements tend to have the largest absolute size, gleaned by applying some selection procedure like that of Benjamini & Hochberg (2015). Previous work on this estimation task proposes the reduction in size of the largest (absolute) sample elements either via shrinkage (like James–Stein) or by subtracting biases estimated using empirical Bayes methodology. We take a novel approach and adapt recent developments by Lee et al. (2016) in post‐selection inference. Adapting and applying their distributional results to our problem post‐selection point and interval estimators for underlying signal sizes are proposed. Simulations suggest that our estimator seems to perform quite well against competitors. Furthermore we prove an upper bound to the so‐called “worst case risk” of our estimator—when combined with the Benjamini–Hochberg selection procedure—and show that it is within a constant multiple of the minimax risk over a rich set of parameter spaces meant to evoke sparsity.The Canadian Journal of Statistics45: 128–148; 2017 © 2017 Statistical Society of CanadaKeywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Exact post-selection inference, with application to the lassoThe Annals of Statistics, 2016
- Empirical Bayes Correction for the Winner's Curse in Genetic Association StudiesGenetic Epidemiology, 2012
- Large-Scale Inference2010
- An empirical Bayes mixture method for effect size and false discovery rate estimationThe Annals of Applied Statistics, 2010
- General maximum likelihood empirical Bayes estimation of normal meansThe Annals of Statistics, 2009
- Estimating Odds Ratios in Genome Scans: An Approximate Conditional Likelihood ApproachThe American Journal of Human Genetics, 2008
- Adapting to unknown sparsity by controlling the false discovery rateThe Annals of Statistics, 2006
- False Discovery Rate–Adjusted Multiple Confidence Intervals for Selected ParametersJournal of the American Statistical Association, 2005
- Needles and straw in haystacks: Empirical Bayes estimates of possibly sparse sequencesThe Annals of Statistics, 2004
- Adapting to Unknown Smoothness via Wavelet ShrinkageJournal of the American Statistical Association, 1995