Alberto Abadie, Mehrdad Ghadiri, Ali Jadbabaie, Mahyar JafariNodeh
arXiv 5 Nov 2025 · Econometrics
arXiv:2511.03236 · PDF · DOI · OpenAlex · Extracted main text
This article introduces a leave-one-out regression adjustment estimator (LOORA) for estimating average treatment effects in randomized controlled trials. The method removes the finite-sample bias of conventional regression adjustment and provides exact variance expressions for LOORA versions of the Horvitz-Thompson and difference-in-means estimators under simple and complete random assignment. Ridge regularization limits the influence of high-leverage observations, improving stability and precision in small samples. In large samples, LOORA attains the asymptotic efficiency of regression-adjusted estimator as characterized by Lin (2013, Annals of Applied Statistics), while remaining exactly unbiased. To construct confidence intervals, we rely on asymptotic variance estimates that treat the estimator as a two-step procedure, accounting for both the regression adjustment and the random assignment stages. Two within-subject experimental applications that provide realistic joint distributions of potential outcomes as ground truth show that LOORA eliminates substantial biases and achieves close-to-nominal confidence interval coverage.
appendix boundary found by none_found · 100% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Lin, Winston (2013) Agnostic notes on regression adjustments to experimental data: reexamining Freedman's critique | 1.000 | 25 | 8 | 100% |
| 2 | Ghadiri, Mehrdad and Arbour, David and Mai, Tung and Musco, Cameron… (2023) Finite population regression adjustment and non-asymptotic guarantees for treatment effect estimation self | 0.874 | 9 | 2 | 100% |
| 3 | Harshaw, Christopher and Sävje, Fredrik and Spielman, Daniel A and Z… (2024) Balancing covariates in randomized experiments with the Gram–Schmidt walk design | 0.874 | 5 | 2 | 100% |
| 4 | Jann Spiess (2025) Optimal estimation when researcher and social preferences are misaligned | 0.874 | 5 | 2 | 100% |
| 5 | Cytrynbaum, Max (2024) Covariate adjustment in stratified experiments | 0.843 | 3 | 3 | 100% |
| 6 | Chang, Haoge and Middleton, Joel A and Aronow, PM (2024) Exact bias correction for linear adjustment of randomized controlled trials | 0.644 | 2 | 2 | 100% |
| 7 | Aronow, Peter M and Middleton, Joel A (2013) A class of unbiased estimators of the average treatment effect in randomized experiments | 0.585 | 3 | 1 | 100% |
| 8 | Allcott, Hunt and Taubinsky, Dmitry (2015) Evaluating behaviorally motivated policy: Experimental evidence from the lightbulb market | 0.511 | 2 | 1 | 100% |
| 9 | Billingsley, P (2012) Probability and Measure | 0.511 | 2 | 1 | 100% |
| 10 | Freedman, David A (2008) On regression adjustments to experimental data | 0.511 | 2 | 1 | 100% |
Showing the top 10 of 32 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | Covariate Adjustment in Randomized Experiments Motivated by Higher-Order Influence Functions | 0.585 | 3 | 1 |
| 2 | Assumption-lean covariate adjustment under covariate adaptive randomization when $p = o (n)$ | 0.405 | 1 | 1 |