2605.04968

Total: 1

#1 Tests for white noise via asymptotically independent U-statistics in high-dimensions [PDF1] [Copy] [Kimi] [REL]

Author: Yuanya Xu

We propose a high-dimensional white noise test that captures serial correlations within and across component series without specifying an alternative model. The test statistic is a U-statistic based on sample autocovariances. Under the null, asymptotic normality is established as $p, T \to \infty$ jointly using martingale difference theory. Our approach imposes no cross-sectional independence assumption, requiring only spectral conditions on $Σ_0$. Theoretically, we link cross-sectional correlations to a graph structure, integrating algebraic and geometric analyses to facilitate the derivation. Simulations confirm reliable size control and satisfactory power across various $(p, T)$ settings.

Subjects: Methodology , Statistics Theory

Publish: 2026-05-06 14:26:50 UTC