Regression-adjusted average treatment effect estimates in stratified randomized experiments
2020-07-21 15:00 — 16:00
Zoom会议号：648 016 53521 会议密码: 693228
Abstract: Researchers often use linear regression to analyse randomized experiments to improve treatment effect estimation by adjusting for imbalances of covariates in the treatment and control groups. Our work offers a randomization-based inference framework for regression adjustment in stratified randomized experiments. Under mild conditions, we re-establish the finite population central limit theorem for a stratified experiment. We prove that both the stratified difference-in-means and the regression-adjusted average treatment effect estimators are consistent and asymptotically normal. The asymptotic variance of the latter is no greater and is typically lesser than that of the former. We also provide conservative variance estimators to construct large-sample confidence intervals for the average treatment effect.
About the Speaker:
Dr. Hanzhong Liu is a tenure-track associate professor at the Center for Statistical Science at Tsinghua University. He was a postdoctoral scholar working with Prof. Bin Yu in the Department of Statistics at UC Berkeley, from 2014 to 2016. Prior to this, he visited UC Berkeley from 2012 to 2014 and received his PhD in statistics from Peking University in 2014, advised by Prof. Bin Yu and Prof. Jinzhu Jia. His research interests lie in statistical theory and methodologies for solving high-dimensional data problems and causal inference, such as, combing bootstrap and sparse modeling method (e.g., Lasso) to construct confidence intervals for parameters in high-dimensional sparse models, and estimating treatment effects using machine learning methods.