{"id":"arxiv/2607.08089","url":"https://arxiv.org/abs/2607.08089","tags":["stat.ME","math.ST"],"score":0,"title":"Bias-Corrected Multiplier Bootstrap Inference for Spectral Edges of Large Covariance Matrices","domain":"stat.ME","source":"arxiv","authors":["Xiucai Ding","Yichen Hu","Jiahui Xie"],"reviews":0,"subject":"stat.ME","summary":"Inference for spectral edges of large covariance matrices is a fundamental problem in high-dimensional statistics. A major difficulty is that the largest non-spiked sample eigenvalues, which serve as natural estimators of the edge, fluctuate on the Tracy--Widom scale. Consequently, valid inference requires accurate centering by the deterministic spectral edge together with a precise scaling constant, both of which are often difficult to estimate in practice under general unknown population covariance structures. In this paper, we propose a bias-corrected multiplier bootstrap procedure for…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-07-09","reactions":{}}