False Discovery Rate Control via Data Splitting
arXiv:2002.08542
Abstract
Selecting relevant features associated with a given response variable is an important issue in many scientific fields. Quantifying quality and uncertainty of a selection result via false discovery rate (FDR) control has been of recent interest. This paper introduces a way of using data-splitting strategies to asymptotically control the FDR while maintaining a high power. For each feature, the method constructs a test statistic by estimating two independent regression coefficients via data splitting. FDR control is achieved by taking advantage of the statistic's property that, for any null feature, its sampling distribution is symmetric about zero. Furthermore, we propose Multiple Data Splitting (MDS) to stabilize the selection result and boost the power. Interestingly and surprisingly, with the FDR still under control, MDS not only helps overcome the power loss caused by sample splitting, but also results in a lower variance of the false discovery proportion (FDP) compared with all other methods in consideration. We prove that the proposed data-splitting methods can asymptotically control the FDR at any designated level for linear and Gaussian graphical models in both low and high dimensions. Through intensive simulation studies and a real-data application, we show that the proposed methods are robust to the unknown distribution of features, easy to implement and computationally efficient, and are often the most powerful ones amongst competitors especially when the signals are weak and the correlations or partial correlations are high among features.
References in corpus (16)
- Adam: A Method for Stochastic Optimization
- Simultaneous analysis of Lasso and Dantzig selector
- On the conditions used to prove oracle results for the Lasso
- Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
- SLOPE - Adaptive variable selection via convex optimization
- Gene Hunting with Knockoffs for Hidden Markov Models
- Robustness of multiple testing procedures against dependence
- On false discovery control under dependence
- DeepPINK: reproducible feature selection in deep neural networks
- Global and Simultaneous Hypothesis Testing for High-Dimensional Logistic Regression Models
- Interpretation of Prediction Models Using the Input Gradient
- A Power and Prediction Analysis for Knockoffs with Lasso Statistics
- Bootstrapping and Sample Splitting For High-Dimensional, Assumption-Free Inference
- Controlling False Discovery Rate Using Gaussian Mirrors
- A Power Analysis of the Conditional Randomization Test and Knockoffs
- GGM knockoff filter: False Discovery Rate Control for Gaussian Graphical Models