Skip to content
Open access

Matching-based estimation of causal quantile exposure response functions for continuous treatments

Oct 2026 · Statistics and computing · Vol 36 · 0 citations · 49 references

Abstract

In this paper we propose novel matching estimators for estimating causal quantile exposure response functions and quantile exposure effects with continuous treatments. Our approach provides a flexible and robust strategy to assess heterogeneous treatment effects across the entire outcome distribution by extending generalized propensity score matching (GPS) to the quantile framework. We provide identification results and establish the asymptotic properties of the derived estimators by using the Bahadur representation for quantiles. The proposed estimation procedure first constructs a matched dataset via GPS matching to adjust for measured confounding, and then fits local kernel reweighted quantile regressions to the matched set. We also derive a consistent estimator of the variance of the matching-type quantile estimators together with a weighted bootstrap procedure for inference. Simulation studies show that the proposed approach outperforms existing weighting-based alternatives in different settings. We apply the proposed method to Medicare claims data for the period 2012-2014 and estimate the causal effect of exposure to PM2.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_{2.5}$$\end{document} on the length of hospital stay for each zip code of the contiguous United States .

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.