Elliptical Regularized Hotelling Testing for High Dimensional Data
Long Feng, Le Zhou, Xiaoyi Wang, [5pt] School of Statistics, Data Science, Nankai University · 2026
Read Elliptical Regularized Hotelling Testing for High Dimensional Data by Long Feng, Le Zhou, Xiaoyi Wang, [5pt] School of Statistics, Data Science, Nankai University (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-25942.txt.
Introduction
Long Feng, Le Zhou, and Xiaoyi Wang build a tougher test for high-dimensional location. Hotelling-style tests ask whether a mean vector sits at a claimed value. Modern data break the classic version: dimensions rival sample size, coordinates depend heavily, and tails run heavy. Their test assumes elliptically symmetric data instead of Gaussian data. A sample spatial median anchors the centre. A spatial-sign covariance matrix, centred at that median, supplies the spread estimate.
Ridge regularization needs a tuning choice. The power-optimal ridge depends on the unknown alternative being tested. Their fix aggregates p-values across a fixed grid of ridge values. A Cauchy combination rule merges them into one analytic p-value. No cross-ridge correlation estimates are needed, thanks to a joint Gaussian limit. Null behaviour and local power are both characterized. Simulations and an application show strong performance under heavy tails and dependence.
Catalog record
| CREATOR | Long Feng, Le Zhou, Xiaoyi Wang, [5pt] School of Statistics, Data Science, Nankai University |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 255 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.25942