Superconsistency of Tests in High Dimensions

Abstract

To assess whether there is some signal in a big database, aggregate tests for the global null hypothesis of no effect are routinely applied in practice before more specialized analysis is carried out. Although a plethora of aggregate tests is available, each test has its strengths but also its blind spots. In a Gaussian sequence model, we study whether it is possible to obtain a test with substantially better consistency properties than the likelihood ratio (i.e., Euclidean norm based) test. We establish an impossibility result, showing that in the high-dimensional framework we consider, the set of alternatives for which a test may improve upon the likelihood ratio test -- that is, its superconsistency points -- is always asymptotically negligible in a relative volume sense.

arXiv commentTypos fixed compared to previous version

References in corpus (13)

• Higher criticism for detecting sparse heterogeneous mixtures
• Sparse Estimators and the Oracle Property, or the Return of Hodges' Estimator
• Global testing under sparse alternatives: ANOVA, multiple comparisons and the higher criticism
• E-values: Calibration, combination, and applications
• Innovated higher criticism for detecting sparse signals in correlated noise
• The sequential rejection principle of familywise error control
• Karl Pearson's meta-analysis revisited
• Feature selection by Higher Criticism thresholding: optimal phase diagram
• Annotated Bibliography of Some Papers on Combining Significances or p-values
• Higher criticism: $p$-values and criticism
• On the volume of the intersection of two $L_p^n$ balls
• Post-selection estimation and testing following aggregated association tests
• How to avoid the zero-power trap in testing for correlation