Adaptive Influence Diagnostics in High-Dimensional Regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Soale, Abdul-Nasah, Lukman, Adewale
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914098564825088
author Soale, Abdul-Nasah
Lukman, Adewale
author_facet Soale, Abdul-Nasah
Lukman, Adewale
contents An adaptive Cook's distance (ACD) for diagnosing influential observations in high-dimensional single-index models with multicollinearity and outlier contamination is proposed. ACD is a model-free technique built on sparse local linear gradients to temper leverage effects. In simulations spanning low- and high-dimensional design settings with strong correlation, ACD based on LASSO (ACD-LASSO) and SCAD (ACD-SCAD) penalties reduced masking and swamping relative to classical Cook's distance and local influence as well as the DF-Model and Case-Weight adjusted solution for LASSO. Trimming points flagged by ACD stabilizes variable selection while preserving core signals. Applications to two datasets--the 1960 US cities pollution study and a high-dimensional riboflavin genomics experiment show consistent gains in selection stability and interpretability.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15618
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Influence Diagnostics in High-Dimensional Regression
Soale, Abdul-Nasah
Lukman, Adewale
Methodology
Applications
Computation
An adaptive Cook's distance (ACD) for diagnosing influential observations in high-dimensional single-index models with multicollinearity and outlier contamination is proposed. ACD is a model-free technique built on sparse local linear gradients to temper leverage effects. In simulations spanning low- and high-dimensional design settings with strong correlation, ACD based on LASSO (ACD-LASSO) and SCAD (ACD-SCAD) penalties reduced masking and swamping relative to classical Cook's distance and local influence as well as the DF-Model and Case-Weight adjusted solution for LASSO. Trimming points flagged by ACD stabilizes variable selection while preserving core signals. Applications to two datasets--the 1960 US cities pollution study and a high-dimensional riboflavin genomics experiment show consistent gains in selection stability and interpretability.
title Adaptive Influence Diagnostics in High-Dimensional Regression
topic Methodology
Applications
Computation
url https://arxiv.org/abs/2510.15618