Stable correlation and robust feature screening

In this paper, we propose a new correlation, called stable correlation, to measure the dependence between two random vectors. The new correlation is well defined without the moment condition and is zero if and only if the two random vectors are independent. We also study its other theoretical properties. Based on the new correlation, we further propose a robust model-free feature screening procedure for ultrahigh dimensional data and establish its sure screening property and rank consistency property without imposing the subexponential or sub-Gaussian tail condition, which is commonly required in the literature of feature screening. We also examine the finite sample performance of the proposed robust feature screening procedure via Monte Carlo simulation studies and illustrate the proposed procedure by a real data example.

Files

  • SciChinaMath.pdf

    size: 1.11 MB | mime_type: application/pdf | date: 2022-07-19 | sha256: becee17

Metadata

Work Title Stable correlation and robust feature screening
Access
Open Access
Creators
  1. Xu Guo
  2. Runze Li
  3. Wanjun Liu
  4. Lixing Zhu
Keyword
  1. Feature screening
  2. Nonlinear dependence
  3. Stable correlation
  4. Sure screening property
License In Copyright (Rights Reserved)
Work Type Article
Publisher
  1. Science China Mathematics
Publication Date April 30, 2021
Publisher Identifier (DOI)
  1. https://doi.org/10.1007/s11425-019-1702-5
Deposited July 19, 2022

Versions

Analytics

Collections

This resource is currently not in any collection.

Work History

Version 1
published

  • Created
  • Added SciChinaMath.pdf
  • Added Creator Xu Guo
  • Added Creator Runze Li
  • Added Creator Wanjun Liu
  • Added Creator Lixing Zhu
  • Published
  • Updated Keyword, Publication Date Show Changes
    Keyword
    • Feature screening, Nonlinear dependence, Stable correlation, Sure screening property
    Publication Date
    • 2022-01-01
    • 2021-04-30