Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data

This paper investigates robust feature screening for ultra-high dimensional data in the presence of outliers and heterogeneity. Considering the susceptibility of likelihood methods to outliers, we propose a Sparse Robust Weighted Expectile Regression (SRoWER) method that combines the <inline-form...

Full description

Saved in:
Bibliographic Details
Main Authors: Xianjun Wu, Pingping Han, Mingqiu Wang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Axioms
Subjects:
Online Access:https://www.mdpi.com/2075-1680/14/5/340
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850127485692280832
author Xianjun Wu
Pingping Han
Mingqiu Wang
author_facet Xianjun Wu
Pingping Han
Mingqiu Wang
author_sort Xianjun Wu
collection DOAJ
description This paper investigates robust feature screening for ultra-high dimensional data in the presence of outliers and heterogeneity. Considering the susceptibility of likelihood methods to outliers, we propose a Sparse Robust Weighted Expectile Regression (SRoWER) method that combines the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>L</mi><mn>2</mn></msub><mi>E</mi></mrow></semantics></math></inline-formula> criterion with expectile regression. By utilizing the IHT algorithm, our method effectively incorporates correlations of covariates and enables joint feature screening. The proposed approach demonstrates robustness against heavy-tailed errors and outliers in data. Simulation studies and a real data analysis are provided to demonstrate the superior performance of the SRoWER method when dealing with outlier-contaminated explanatory variables and/or heavy-tailed error distributions.
format Article
id doaj-art-c9a7d3ae2a354469ae3e5db5a1db7ad8
institution OA Journals
issn 2075-1680
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Axioms
spelling doaj-art-c9a7d3ae2a354469ae3e5db5a1db7ad82025-08-20T02:33:39ZengMDPI AGAxioms2075-16802025-04-0114534010.3390/axioms14050340Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional DataXianjun Wu0Pingping Han1Mingqiu Wang2School of Statistics and Data Science, Qufu Normal University, Qufu 273100, ChinaSchool of Statistics and Data Science, Qufu Normal University, Qufu 273100, ChinaSchool of Statistics and Data Science, Qufu Normal University, Qufu 273100, ChinaThis paper investigates robust feature screening for ultra-high dimensional data in the presence of outliers and heterogeneity. Considering the susceptibility of likelihood methods to outliers, we propose a Sparse Robust Weighted Expectile Regression (SRoWER) method that combines the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>L</mi><mn>2</mn></msub><mi>E</mi></mrow></semantics></math></inline-formula> criterion with expectile regression. By utilizing the IHT algorithm, our method effectively incorporates correlations of covariates and enables joint feature screening. The proposed approach demonstrates robustness against heavy-tailed errors and outliers in data. Simulation studies and a real data analysis are provided to demonstrate the superior performance of the SRoWER method when dealing with outlier-contaminated explanatory variables and/or heavy-tailed error distributions.https://www.mdpi.com/2075-1680/14/5/340asymmetric least squaresfeature screeningheteroscedasticityrobust regressionultra-high dimensional data
spellingShingle Xianjun Wu
Pingping Han
Mingqiu Wang
Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data
Axioms
asymmetric least squares
feature screening
heteroscedasticity
robust regression
ultra-high dimensional data
title Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data
title_full Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data
title_fullStr Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data
title_full_unstemmed Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data
title_short Sparse Robust Weighted Expectile Screening for Ultra-High-Dimensional Data
title_sort sparse robust weighted expectile screening for ultra high dimensional data
topic asymmetric least squares
feature screening
heteroscedasticity
robust regression
ultra-high dimensional data
url https://www.mdpi.com/2075-1680/14/5/340
work_keys_str_mv AT xianjunwu sparserobustweightedexpectilescreeningforultrahighdimensionaldata
AT pingpinghan sparserobustweightedexpectilescreeningforultrahighdimensionaldata
AT mingqiuwang sparserobustweightedexpectilescreeningforultrahighdimensionaldata