A0481
Title: Regression adjustment in randomized controlled trials with many covariates
Authors: Harold Chiang - University of Wisconsin-Madison (United States)
Yukitoshi Matsushita - Hitotsubashi University (Japan) [presenting]
Taisuke Otsu - London School of Economics (United Kingdom)
Abstract: The proposed method estimates and predicts average treatment effects in randomized controlled trials when researchers observe potentially many covariates. By employing Neyman's finite population perspective, a bias-corrected regression adjustment estimator using cross-fitting is proposed, and it is shown that the proposed estimator has favourable properties over existing alternatives. For inference, the first and second-order terms are derived in the stochastic component of the regression adjustment estimators, higher-order properties of the existing inference methods are studied, and a bias-corrected version of the HC3 standard error is proposed. Simulation studies show our cross-fitted estimator, combined with the bias-corrected HC3, delivers precise point estimates and robust size controls over a wide range of DGPs. To illustrate, the proposed methods are applied to real datasets on randomized experiments of incentives and services for college achievement following previous researchers.
