Unbiased Regression-Adjusted Estimation of Average
Treatment Effects in Randomized Controlled Trials
Name
jafarinodeh-mahyarjn-sm-idss-26-thesis.pdf
Size
730.28 KB
Format
Adobe PDF
Checksum (MD5)
d434605561001f489f3683ceb0246692
Author(s)
JafariNodeh, Mahyar
Advisor(s)
Jadbabaie, Ali
Abadie, Alberto
Date Issued
February 2026
Publisher
Massachusetts Institute of Technology
Abstract
This article introduces a leave-one-out regression adjustment (LOORA) for estimating average treatment effects in randomized controlled trials. The method removes the finitesample bias of conventional regression adjustment and provides exact variance formulas for LOORA versions of the Horvitz–Thompson and difference-in-means estimators under simple and complete random assignment. Ridge regularization curbs the influence of high-leverage observations, improving stability and precision in small samples. In large samples, LOORA matches the variance of the regression-adjusted estimator in Lin 2013, while remaining exactly unbiased. Two within-subject experimental applications, each providing realistic joint distributions of potential outcomes as ground truth, show that LOORA removes substantial biases and achieves close-to-nominal confidence interval coverage.
MIT Department
Massachusetts Institute of Technology. Institute for Data, Systems, and Society
Terms of Use
In Copyright - Educational Use Permitted
Copyright retained by author(s)
Persistent DSpace Link
