Valid and efficient inference for nonparametric variable importance in two-phase studies
Biometrics · 2025
DOI: 10.1093/biomtc/ujaf095 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biometrics, 2025. Verfasst von Guorong Dai und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics and Data Science, School of Management, Fudan University, Shanghai 200433, China. Hauptthemen laut MEDLINE: Models, Statistical, Biometry. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 20 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: National Natural Science Foundation of China, NCI NIH HHS, NIGMS NIH HHS.
Abstract (Originalsprache)
We consider a common nonparametric regression setting, where the data consist of a response variable Y, some easily obtainable covariates $\mathbf {X}$, and a set of costly covariates $\mathbf {Z}$. Before establishing predictive models for Y, a natural question arises: Is it worthwhile to include $\mathbf {Z}$ as predictors, given the additional cost of collecting data on $\mathbf {Z}$ for both training the models and predicting Y for future individuals? Therefore, we aim to conduct preliminary investigations to infer importance of $\mathbf {Z}$ in predicting Y in the presence of $\mathbf {X}$. To achieve this goal, we propose a nonparametric variable importance measure for $\mathbf {Z}$. It is defined as a parameter that aggregates maximum potential contributions of $\mathbf {Z}$ in single or multiple predictive models, with contributions quantified by general loss functions. Considering two-phase data that provide a large number of observations for $(Y,\mathbf {X})$ with the expensive $\mathbf {Z}$ measured only in a small subsample, we develop a novel approach to infer the proposed importance measure, accommodating missingness of $\mathbf {Z}$ in the sample by substituting functions of $(Y,\mathbf {X})$ for each individual’s contribution to the predictive loss of models involving $\mathbf {Z}$. Our approach attains unified and efficient inference regardless of whether $\mathbf {Z}$ makes zero or positive contribution to predicting Y, a desirable yet surprising property owing to data incompleteness. As intermediate steps of our theoretical development, we establish novel results in two relevant research areas, semi-supervised inference and two-phase nonparametric estimation. Numerical results from both simulated and real data demonstrate superior performance of our approach.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Statistics and Data Science, School of Management, Fudan University, Shanghai 200433, China.
- Department of Statistics, Texas A&M University, College Station, TX 77840, United States.
- Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, United States.
ORCID-Profile: Guorong Dai, Raymond J Carroll
Förderung
- National Natural Science Foundation of China
- NCI NIH HHS
- NIGMS NIH HHS
- NIH HHS
- NHLBI NIH HHS
- National Institutes of Health
- National Cancer Institute
Zitieren
APA
Dai, G., Carroll, R. J., & Chen, J. (2025). Valid and efficient inference for nonparametric variable importance in two-phase studies. Biometrics. https://doi.org/10.1093/biomtc/ujaf095
Vancouver
Dai G, Carroll RJ, Chen J. Valid and efficient inference for nonparametric variable importance in two-phase studies. Biometrics. 2025. doi:10.1093/biomtc/ujaf095
BibTeX
@article{Dai2025,
author = {Dai, Guorong and Carroll, Raymond J and Chen, Jinbo},
title = {Valid and efficient inference for nonparametric variable importance in two-phase studies},
journal = {Biometrics},
year = {2025},
doi = {10.1093/biomtc/ujaf095}
}
EUVIMED