Density and treatment effect estimation under covariate-adaptive randomization with heavy-tailed outcomes
Biometrics · 2026
DOI: 10.1093/biomtc/ujag159 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biometrics, 2026. Verfasst von Hongzi Li und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics and Data Science, Tsinghua University, Beijing, 100084, China. Hauptthemen laut MEDLINE: Models, Statistical, Randomized Controlled Trials as Topic. Die Arbeit nennt 24 Literaturangaben. Gefördert von: Beijing Natural Science Foundation, National Natural Science Foundation of China, Shenzhen Science and Technology.
Abstract (Originalsprache)
Randomized experiments are the gold standard for investigating causal relationships, with comparisons of potential outcomes under different treatment groups used to estimate treatment effects. However, outcomes with heavy-tailed distributions pose significant challenges to traditional causal inference approaches. While recent studies have explored these issues under simple randomization, their application in more complex randomization designs, such as stratified randomization or covariate-adaptive randomization, has not been adequately addressed. To fill the gap, we first investigate the performance of nonparametric kernel density estimation methods under covariate-adaptive randomization, thereby establishing theoretical guarantees for treatment effect estimators based on the estimated densities. Second, we demonstrate the application of our density estimation framework to estimate the density treatment effect and overall quantile treatment effect, deriving the consistency and asymptotic normality of the estimators. For the overall quantile treatment effect, we show that the existing variance estimator for the influence function-based M-estimator tends to overestimate the asymptotic variance, especially under more balanced designs, and lacks universal applicability across randomization methods. To remedy this, we introduce a novel stratified transformed difference-in-means estimator to enhance efficiency and propose a universally applicable variance estimator to facilitate valid inferences. Numerical results demonstrate the effectiveness of the proposed methods in finite samples.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Statistics and Data Science, Tsinghua University, Beijing, 100084, China.
- Institute of Statistics and Big Data, Renmin University of China, Beijing, 100872, China.
- School of Economics and Management, Beihang University, Beijing, 100191, China.
ORCID-Profile: Wei Ma, Hanzhong Liu
Förderung
- Beijing Natural Science Foundation
- National Natural Science Foundation of China
- Shenzhen Science and Technology
Zitieren
APA
Li, H., Ma, W., Ma, Y., & Liu, H. (2026). Density and treatment effect estimation under covariate-adaptive randomization with heavy-tailed outcomes. Biometrics. https://doi.org/10.1093/biomtc/ujag159
Vancouver
Li H, Ma W, Ma Y, Liu H. Density and treatment effect estimation under covariate-adaptive randomization with heavy-tailed outcomes. Biometrics. 2026. doi:10.1093/biomtc/ujag159
BibTeX
@article{Li2026,
author = {Li, Hongzi and Ma, Wei and Ma, Yingying and Liu, Hanzhong},
title = {Density and treatment effect estimation under covariate-adaptive randomization with heavy-tailed outcomes},
journal = {Biometrics},
year = {2026},
doi = {10.1093/biomtc/ujag159}
}
EUVIMED