Optimality of the max test for detecting sparse signals with Gaussian or heavier tail

TL;DR

This paper proves the optimality of the max test for detecting sparse signals in high-dimensional data with Gaussian or heavier tails, extending previous results to more general non-null mean distributions.

Contribution

It establishes the max test's optimal detection boundary in the weak, sparse regime for distributions with tails no lighter than Gaussian, and analyzes the limitations of the higher criticism test.

Findings

Max test is optimal in the weak, sparse regime for Gaussian or heavier tails.

Higher criticism can have very low power with polynomial tail distributions.

Theoretical and simulation results confirm the detection boundaries.

Abstract

A fundamental problem in high-dimensional testing is that of global null testing: testing whether the null holds simultaneously in all of $n$ hypotheses. The max test, which uses the smallest of the $n$ marginal p-values as its test statistic, enjoys widespread popularity for its simplicity and robustness. However, its theoretical performance relative to other tests has been called into question. In the Gaussian sequence version of the global testing problem, Donoho and Jin (2004) discovered a so-called "weak, sparse" asymptotic regime in which the higher criticism and Berk-Jones tests achieve a better detection boundary than the max test when all of the nonzero signal strengths are identical. We study a more general model in which the non-null means are drawn from a generic distribution, and show that the detection boundary for the max test is optimal in the "weak, sparse" regime, provided that the distribution's tail is no lighter than Gaussian. Further, we show theoretically and in simulation that the modified higher criticism of Donoho and Jin (2004) can have very low power when the distribution of non-null means has a polynomial tail.