Trade-off between predictive performance and FDR control for high-dimensional Gaussian model selection

TL;DR

This paper introduces a new calibration method for penalized least-squares in high-dimensional Gaussian regression that balances predictive accuracy and false discovery rate, validated through simulations.

Contribution

It proposes a novel hyperparameter calibration algorithm that controls both prediction error and FDR, applicable even without variable ordering.

Findings

The calibration algorithm effectively balances FDR and prediction risk.

The method outperforms existing variable selection procedures in simulations.

Extension to unordered variables broadens applicability.

Abstract

In the context of high-dimensional Gaussian linear regression for ordered variables, we study the variable selection procedure via the minimization of the penalized least-squares criterion. We focus on model selection where the penalty function depends on an unknown multiplicative constant commonly calibrated for prediction. We propose a new proper calibration of this hyperparameter to simultaneously control predictive risk and false discovery rate. We obtain non-asymptotic bounds on the False Discovery Rate with respect to the hyperparameter and we provide an algorithm to calibrate it. This algorithm is based on quantities that can typically be observed in real data applications. The algorithm is validated in an extensive simulation study and is compared with several existing variable selection procedures. Finally, we study an extension of our approach to the case in which an ordering of the variables is not available.