Abstract:Rare and Weak models for multiple hypothesis testing assume that only a small proportion of the tested hypotheses concern non-null effects and the individual effects are only moderately large, so they generally do not stand out individually, for example in a Bonferroni analysis. Such models have been studied in quite a few settings, for example in some cases studies focused on an underlying Gaussian means model for the hypotheses being tested; in some others, Poisson and Binomial. Such seemingly different models have asymptotically the following common structure. Summarizing the evidence of individual tests by the negative logarithm of its P-value, the model is asymptotically equivalent to a situation in which most negative log P-values have a standard exponential distribution but a small fraction of the P-values might have an alternative distribution which is approximately noncentral chisquared on one degree of freedom. This log-chisquared approximation is different from the log-normal approximation of Bahadur which is unsuitable for analyzing Rare and Weak multiple testing models. We characterize the asymptotic performance of global tests combining asymptotic log-chisquared P-values in terms of the chisquared mixture parameters: the scaling parameter controlling heteroscedasticity, the non-centrality parameter describing the effect size whenever it exists, and the parameter controlling the rarity of the non-null effects. In a phase space involving the last two parameters, we derive a region where all tests are asymptotically powerless. Outside of this region, the Berk-Jones and the Higher Criticism tests have maximal power. Inference techniques based on the minimal P-value, false-discovery rate controlling, and Fisher's combination test have sub-optimal asymptotic phase diagrams.

Statistical hypothesis testing by weak-value amplification: Proposal and evaluation

Weak value amplification: a view from quantum estimation theory that highlights what it is and what isn't

Uncertainty of Weak Measurement and Merit of Amplification

Progress and Perspectives on Weak-value Amplification

Experimental investigation of high-efficiency weak-value amplification of nonunitary evolution

Amplification Limit of Weak Measurements: A Variational Approach

Weak values could reveal the hidden effects of quantum interactions

Effect of Spectral Resolution on Weak-Value Based Parameter Estimation

On Amplification by Weak Measurement

Amplification Limit of Weak Measurement

Autocorrelative Weak-Value Amplification and Simulating the Protocol under Strong Gaussian Noise

Approaching Quantum-Limited Metrology with Imperfect Detectors by Using Weak-Value Amplification

Auto-correlative weak-value amplification under strong noise background

Pragmatic Implementation with Non-Gaussian Devices of Noise-Limited Weak Value Amplification

Ultra-small phase estimation via weak measurement with postselection: A comparison of joint weak measurement and weak value amplification

A NEW PERSPECTIVE ON ROBUST M-ESTIMATION: FINITE SAMPLE THEORY AND APPLICATIONS TO DEPENDENCE-ADJUSTED MULTIPLE TESTING

The systematic error and the anomaly in the sensitivity and the probability of post-selection raised from the plural weak value

Near-deterministic weak-value metrology via collective non-linearity

Experimental probe of weak value amplification and geometric phase through the complex zeroes of the response function

Unification of Rare/Weak Detection Models using Moderate Deviations Analysis and Log-Chisquared P-values

Weak Value Amplification Via Second-Order Correlated Technique