Analytically Calibrated Upper L-Statistics for High-Dimensional Mean Change-Point Testing
Abstract
This paper develops upper L-statistic methods for testing and localizing a high-dimensional mean change. At each candidate split, the proposed statistics rank the squared studentized CUSUM values across coordinates and aggregate either a fixed number or a fixed proportion of the largest entries. This construction connects max-type and sum-type procedures and provides sensitivity to a wide range of sparsity regimes. We establish asymptotic null distributions under an independent-component representation allowing weak cross-sectional dependence. The limiting laws are Poisson tangent processes for fixed ranks and Gaussian-process or Gumbel laws for proportional ranks. To improve finite-sample performance, we develop analytic calibrations for marginal variance estimation, cross-sectional dependence, extreme clustering, and skewness, and establish their first-order validity. We combine calibrated p-values from a matched pair of fixed- and proportional-order statistics using the Cauchy combination rule to adapt to unknown sparsity. For localization, the estimated change point is selected from the component scan with the smallest p-value. Simulation studies demonstrate generally satisfactory size calibration, competitive detection power, and effective localization across different sparsity levels. An application to semiconductor manufacturing data further illustrates the practical utility of the proposed methods.
Related
- How to Run Statistics over LLM Judges and Trust the Results: Calibrated Inference for Small-Sample AI Evaluation with evalstats
- DOMIC: Provably Calibrated Detection of Dependence-Structure Change Points via Density-Operator Mutual Information
- Boot-AVT: Asymptotically Anytime-Valid Testing with Degenerate U-Statistics via Bootstrap
- Risk-Calibrated Balancing for High-Dimensional Causal Extrapolation
- Max-Test-Calibrated Stein Shrinkage with Honest Submodel Selection in Ultra-High-Dimensional Regression
BibTeX
@misc{yan2026analytically,
title = {{Analytically Calibrated Upper L-Statistics for High-Dimensional Mean Change-Point Testing}},
author = {Guowei Yan and Dan Zhuang and Long Feng},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.04364},
eprint = {2611.04364},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.04364},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Analytically Calibrated Upper L-Statistics for High-Dimensional Mean Change-Point Testing
AU - Guowei Yan
AU - Dan Zhuang
AU - Long Feng
PY - 2026
DA - 2026/11/03
PB - arXiv
DO - 10.48550/arxiv.2611.04364
UR - https://arxiv.org/abs/2611.04364
AB - This paper develops upper L-statistic methods for testing and localizing a high-dimensional mean change. At each candidate split, the proposed statistics rank the squared studentized CUSUM values across coordinates and aggregate either a fixed number or a fixed proportion of the largest entries. This construction connects max-type and sum-type procedures and provides sensitivity to a wide range of sparsity regimes. We establish asymptotic null distributions under an independent-component representation allowing weak cross-sectional dependence. The limiting laws are Poisson tangent processes for fixed ranks and Gaussian-process or Gumbel laws for proportional ranks. To improve finite-sample performance, we develop analytic calibrations for marginal variance estimation, cross-sectional dependence, extreme clustering, and skewness, and establish their first-order validity. We combine calibrated p-values from a matched pair of fixed- and proportional-order statistics using the Cauchy combination rule to adapt to unknown sparsity. For localization, the estimated change point is selected from the component scan with the smallest p-value. Simulation studies demonstrate generally satisfactory size calibration, competitive detection power, and effective localization across different sparsity levels. An application to semiconductor manufacturing data further illustrates the practical utility of the proposed methods.
ER -