{"id":"arxiv/2605.04968","url":"https://arxiv.org/abs/2605.04968","tags":["stat.ME","math.ST"],"score":0,"title":"Tests for white noise via asymptotically independent U-statistics in high-dimensions","domain":"stat.ME","source":"arxiv","authors":["Yuanya Xu"],"reviews":0,"subject":"stat.ME","summary":"Existing tests for high-dimensional white noise are tailored to the two ends of the sparsity spectrum: maximum-type statistics excel against a few dominant correlations and sum-type statistics against many weak ones, leaving the intermediate regime without a systematic tool. We propose a family of $U$-statistics built from the entry-wise $\\ell_a$-norms of the sample autocovariance matrices, in which the even order $a$ interpolates between the sum-type ($a=2$) and maximum-type ($a\\to\\infty$) endpoints. As $p$ and $T$ diverge jointly, martingale difference theory yields joint asymptotic…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-05-06","reactions":{}}