讲座题目：A Data Fusion Method for Quantile Treatment Effects

主讲人：朱仲义教授 复旦大学

讲座时间：2022年6月15日下午14点-15点

讲座地点：腾讯会议号：219 268 501

主讲人简介：

朱仲义，复旦大学统计系教授，博士生导师；曾任中国概率统计学会第八、九届副理事长,国际著名杂志《Statistica Sinica》副主编；《应用概率统计》和《数理统计与管理》杂志编委，中国统计教材编审委员会委员；现为 Elected ISI Member（国际统计学会推选会员），《中国科学:数学》杂志编委。研究方向包括保险精算、纵向数据(面板数据)模型、分位数回归模型等统计推断问题研究。目前主持国家自然科学基金重大项目子项目一项，重点项目子项目一项以及面上项目一项，已主持完成国家自然科学基金四项、国家社会科学基金一项。发表学术论文100多篇，包括在国际四大统计顶级刊物等SCI论文六十多篇，并获教育部自然科学二等奖一次。

主讲内容：

With the increasing availability of datasets, developing data fusion methods to leverage the strengths of different types of data to draw causal effects is of great practical importance to many scientific fields. In this paper, we consider estimating the quantile treatment effects with small validation data with fully-observed confounders and large auxiliary data with unmeasured confounders. We propose a fused quantile treatment effects estimator (FQTE) by integrating the information from two datasets based on doubly robust estimating functions. We allow for the misspecification of the models on the dataset with unmeasured confounders. Under mild conditions, we show that the proposed FQTE is asymptotically normal and more efficient than the initial QTE estimator using the validation data solely. By establishing the asymptotic linear forms of related estimators, convenient methods for covariance estimation are provided to make our method easy to implement. Simulation studies demonstrate the empirical validity and improved efficiency of our fused estimators. We illustrate the proposed method with an application.