EigenPrism: inference for high dimensional signal-to-noise ratios

• 成果类型: Article
• 署名作者: Janson, Lucas; Barber, Rina Foygel; Candes, Emmanuel
• 署名单位: Stanford University; University of Chicago
• 刊物名称: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
• ISSN/ISSBN: 1369-7412
• DOI: 10.1111/rssb.12203
• 发表日期: 2017
• 页码: 1037-1065
• 关键词: wide association analysis; genome-wide; heritability; regression; variance; asymptotics; selection; MODEL

摘要

Consider the following three important problems in statistical inference: constructing confidence intervals for the error of a high dimensional (p>n) regression estimator, the linear regression noise level and the genetic signal-to-noise ratio of a continuous-valued trait ( related to the heritability). All three problems turn out to be closely related to the little-studied problem of performing inference on the (l)2-norm of the signal in high dimensional linear regression. We derive a novel procedure for this, which is asymptotically correct when the covariates are multivariate Gaussian and produces valid confidence intervals in finite samples as well. The procedure, called EigenPrism, is computationally fast and makes no assumptions on coefficient sparsity or knowledge of the noise level. We investigate the width of the EigenPrism confidence intervals, including a comparison with a Bayesian setting in which our interval is just 5% wider than the Bayes credible interval. We are then able to unify the three aforementioned problems by showing that EigenPrism with only minor modifications can make important contributions to all three. We also investigate the robustness of coverage and find that the method applies in practice and in finite samples much more widely than just the case of multivariate Gaussian covariates. Finally, we apply EigenPrism to a genetic data set to estimate the genetic signal-to-noise ratio for a number of continuous phenotypes.