A central limit theorem for the Benjamini-Hochberg false discovery proportion under a factor model
From MaRDI portal
Publication:6178583
DOI10.3150/23-BEJ1615arXiv2104.08687OpenAlexW3152555339MaRDI QIDQ6178583FDOQ6178583
Publication date: 16 January 2024
Published in: Bernoulli (Search for Journal in Brave)
Abstract: The Benjamini-Hochberg (BH) procedure remains widely popular despite having limited theoretical guarantees in the commonly encountered scenario of correlated test statistics. Of particular concern is the possibility that the method could exhibit bursty behavior, meaning that it might typically yield no false discoveries while occasionally yielding both a large number of false discoveries and a false discovery proportion (FDP) that far exceeds its own well controlled mean. In this paper, we investigate which test statistic correlation structures lead to bursty behavior and which ones lead to well controlled FDPs. To this end, we develop a central limit theorem for the FDP in a multiple testing setup where the test statistic correlations can be either short-range or long-range as well as either weak or strong. The theorem and our simulations from a data-driven factor model suggest that the BH procedure exhibits severe burstiness when the test statistics have many strong, long-range correlations, but does not otherwise.
Full work available at URL: https://arxiv.org/abs/2104.08687
multiple hypothesis testingfunctional central limit theoremfunctional delta methodsimes lineempirical cumulative distribution function
Cites Work
- Multiple hypothesis testing adjusted for latent variables, with an application to the AGEMAP gene expression data
- A Factor Model Approach to Multiple Testing Under Dependence
- FarmTest: Factor-Adjusted Robust Multiple Testing With Approximate False Discovery Control
- Asymptotic Statistics
- Title not available (Why is that?)
- Estimating False Discovery Proportion Under Arbitrary Covariance Dependence
- The control of the false discovery rate in multiple testing under dependency.
- A stochastic process approach to false discovery control.
- The effect of correlation in false discovery rate estimation
- Correlation and Large-Scale Simultaneous Significance Testing
- Statistical significance for genomewide studies
- Bi-cross-validation for factor analysis
- Strong Control, Conservative Point Estimation and Simultaneous Conservative Consistency of False Discovery Rates: A Unified Approach
- Variance of the Number of False Discoveries
- More powerful control of the false discovery rate under dependence
- Mixing properties of ARMA processes
- Control of generalized error rates in multiple testing
- Asymptotic properties of false discovery rate controlling procedures under independence
- Dependency and false discovery rate: asymptotics
- Some Results on the Control of the False Discovery Rate under Dependence
- Controlling the number of false discoveries: application to high-dimensional genomic data
- Title not available (Why is that?)
- New procedures controlling the false discovery proportion via Romano-Wolf's heuristic
- An Introduction to Functional Central Limit Theorems for Dependent Stochastic Processes
- Stationarity, Mixing, Distributional Properties and Moments of GARCH(p, q)–Processes
- Multiple hypotheses testing and expected number of type I errors
- Asymptotic FDR control under weak dependence: a counterexample
- On the False Discovery Rate and Expected Type I Errors
- A central limit theorem for triangular arrays of weakly dependent random variables, with applications in statistics
- Multiple testing via \(\mathrm{FDR}_L\) for large-scale imaging data
- On the false discovery proportion convergence under Gaussian equi-correlation
- The Empirical Distribution of a Large Number of Correlated Normal Variables
- On the performance of FDR control: constraints and a partial solution
- Strong consistency and asymptotic normality for quantities related to the Benjamini-Hochberg false discovery rate procedure
- Confounder adjustment in multiple hypothesis testing
- Detecting multiple replicating signals using adaptive filtering procedures
- On empirical distribution function of high-dimensional Gaussian vector components with an application to multiple testing
- Linear Models and Time-Series Analysis
- Conditional calibration for false discovery rate control under dependence
This page was built for publication: A central limit theorem for the Benjamini-Hochberg false discovery proportion under a factor model
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6178583)
