Testing in high-dimensional spiked models

TL;DR

This paper analyzes high-dimensional multivariate tests involving eigenvalues of Wishart matrices, deriving asymptotic power envelopes and revealing Gaussian process limits for the likelihood ratios under sub-critical spikes.

Contribution

It provides a unified framework for testing in high-dimensional spiked models, including new asymptotic distributions and power bounds for eigenvalue-based tests.

Findings

Log likelihood ratios converge to Gaussian processes with logarithmic correlation.

Asymptotic power envelopes for spike detection are derived.

Unified approach applies across multiple classical multivariate problems.

Abstract

We consider the five classes of multivariate statistical problems identified by James (1964), which together cover much of classical multivariate analysis, plus a simpler limiting case, symmetric matrix denoising. Each of James' problems involves the eigenvalues of $E^{-1}H$ where $H$ and $E$ are proportional to high dimensional Wishart matrices. Under the null hypothesis, both Wisharts are central with identity covariance. Under the alternative, the non-centrality or the covariance parameter of $H$ has a single eigenvalue, a spike, that stands alone. When the spike is smaller than a case-specific phase transition threshold, none of the sample eigenvalues separate from the bulk, making the testing problem challenging. Using a unified strategy for the six cases, we show that the log likelihood ratio processes parameterized by the value of the sub-critical spike converge to Gaussian processes with logarithmic correlation. We then derive asymptotic power envelopes for tests for the presence of a spike.