Model-free and Distributionally Robust Feature Screening with False Discovery Control for High-Dimensional Heterogeneous Data
Abstract
In this paper, we propose a model-free feature screening framework tailored for high-dimensional and heterogeneous datasets, based on a novel distributionally robust dependence measure termed Copula Divergence. The proposed screening method, named CD-Screen, addresses critical limitations of existing feature screening methods, such as restrictive modeling assumptions and sensitivity to heterogeneous feature distributions. CD-Screen ranks features according to their Copula Divergence without relying on a specific regression model or distributional assumptions. Additionally, we introduce CD-FDR, a data-driven procedure to control false discoveries, ensuring accurate and efficient feature selection. Theoretical analyses establish the sure screening and rank consistency properties of CD-Screen, along with asymptotic control of the false discovery rate by CD-FDR. Extensive simulation studies demonstrate the superior performance of our methods compared to traditional screening approaches across diverse scenarios. Furthermore, a real data analysis of the relationship between stock returns and inflation in the United States {illustrates the practical use of our method and provides descriptive evidence on} sector-specific responses to economic changes.
Related
- Transfer Learning with Heterogeneous Feature Spaces in Linear Regression
- Feature Space Selection and Heterogeneous Effect Estimation for Blood-Brain Barrier Permeability: A Random Forest to the Generalized Random Forest Pipeline
- On the Duality between Feature and Sample Screening
- What Do Sparse Autoencoder Feature Discovery Scores Identify? From Activations to Interventions
- Interpretable Feature Discovery from Unstructured Data Using LLMs and Statistical Choice Models
BibTeX
@misc{cheng2026modelfree,
title = {{Model-free and Distributionally Robust Feature Screening with False Discovery Control for High-Dimensional Heterogeneous Data}},
author = {Cong Cheng and Runze Li and Yuan Ke},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.23635},
eprint = {2609.23635},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.23635},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Model-free and Distributionally Robust Feature Screening with False Discovery Control for High-Dimensional Heterogeneous Data
AU - Cong Cheng
AU - Runze Li
AU - Yuan Ke
PY - 2026
DA - 2026/09/20
PB - arXiv
DO - 10.48550/arxiv.2609.23635
UR - https://arxiv.org/abs/2609.23635
AB - In this paper, we propose a model-free feature screening framework tailored for high-dimensional and heterogeneous datasets, based on a novel distributionally robust dependence measure termed Copula Divergence. The proposed screening method, named CD-Screen, addresses critical limitations of existing feature screening methods, such as restrictive modeling assumptions and sensitivity to heterogeneous feature distributions. CD-Screen ranks features according to their Copula Divergence without relying on a specific regression model or distributional assumptions. Additionally, we introduce CD-FDR, a data-driven procedure to control false discoveries, ensuring accurate and efficient feature selection. Theoretical analyses establish the sure screening and rank consistency properties of CD-Screen, along with asymptotic control of the false discovery rate by CD-FDR. Extensive simulation studies demonstrate the superior performance of our methods compared to traditional screening approaches across diverse scenarios. Furthermore, a real data analysis of the relationship between stock returns and inflation in the United States {illustrates the practical use of our method and provides descriptive evidence on} sector-specific responses to economic changes.
ER -