Events

Department Seminars

• 

  Wednesday, December 10, 2025

  1:30 pm Central Time online

  Causal Inference on Quantile Dose-response Functions via Local ReLU Least Squares Weighting

  We are pleased to host a virtual talk by Shujie Ma, PhD, titled "Causal Inference on Quantile Dose-response Functions via Local ReLU Least Squares Weighting." Estimating quantile dose–response functions is crucial in understanding how treatment effects vary across different exposure levels in many scientific and biomedical applications. Dr. Ma will present a new local ReLU least-squares weighting method for this problem that estimates the weighting function directly, rather than relying on inverse propensity weighting (IPW). The proposed approach integrates two-layer ReLU networks to manage high-dimensional covariates and local kernel smoothing to handle continuous treatments, offering a scalable and computationally efficient framework for quantile dose–response estimation. Compared with conventional IPW, the method improves robustness and numerical stability while avoiding density estimation and inversion steps. On the theoretical side, Dr. Ma will introduce a mixed fractional Sobolev function class and show that two-layer ReLU networks can break the curse of dimensionality when the weighting function belongs to this class. She will further establish convergence rates for the learned weights and asymptotic normality for the proposed estimator, and propose a multiplier bootstrap method to construct confidence bands for quantile dose-response functions. Simulations demonstrate strong finite-sample performance and practical tuning strategies.

  Shujie Ma is a professor and graduate advisor in the Department of Statistics at the University of California, Riverside. She received her PhD from Michigan State University in 2011. Her current research interests include precision medicine, causal inference, deep learning theory, and applications for large-scale data, network analysis, and semiparametric inference. She is an elected Fellow of the American Statistical Association (ASA), Institute of Mathematical Statistics, and International Statistical Institute. Dr. Ma has served on numerous editorial boards, including the Annals of Statistics, the Journal of the American Statistical Association, the Journal of Computational and Graphical Statistics, and the Journal of Business & Economic Statistics, among others. She is currently co-editor of ASA Discoveries, the ASA’s open-access journal for research across statistics, data science, and AI.

   

   

  Request access