Regression-adjusted average treatment effect estimates in stratified randomized experiments

DOI10.1093/BIOMET/ASAA038zbMATH Open1457.62046arXiv1908.01628OpenAlexW3034992734MaRDI QIDQ5145706FDOQ5145706

Authors: Hanzhong Liu, Yuehan Yang 

Publication date: 21 January 2021

Published in: Biometrika (Search for Journal in Brave)

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.

Full work available at URL: https://arxiv.org/abs/1908.01628

Recommendations

• Regression-assisted inference for the average treatment effect in paired experiments
• Rerandomization and regression adjustment
• On regression adjustments in experiments with several treatments
• Revisiting regression adjustment in experiments with heterogeneous treatment effects
• Lasso adjustments of treatment effect estimates in randomized experiments

zbMATH Keywords

randomized experimentsrandomization-based inferencestratified samplingblockingrandomized block design

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Sampling theory, sample surveys (62D05) Statistical block designs (62K10)

Cited In (23)

• No star is good news: a unified look at rerandomization based on \(p\)-values from covariate balance tests
• Design-Based Ratio Estimators and Central Limit Theorems for Clustered, Blocked RCTs
• Regression analysis for covariate-adaptive randomization: a robust and efficient inference perspective
• Covariate-adjusted Fisher randomization tests for the average treatment effect
• Reconciling design-based and model-based causal inferences for split-plot experiments
• Regression-based multiple treatment effect estimation under covariate-adaptive randomization
• Bias of the regression estimator for experiments using clustered random assignment
• The Generalized Oaxaca-Blinder Estimator
• Model-Robust Inference for Clinical Trials that Improve Precision by Stratified Randomization and Covariate Adjustment
• Comment on ‘Inference after covariate-adaptive randomisation: aspects of methodology and theory’
• Model-Assisted Complier Average Treatment Effect Estimates in Randomized Experiments with Noncompliance
• Adjusting Treatment Effect Estimates by Post-Stratification in Randomized Experiments
• Pair-Switching Rerandomization
• Regression-adjusted estimation of quantile treatment effects under covariate-adaptive randomizations
• A unified analysis of regression adjustment in randomized experiments
• A joint estimation for the high-dimensional regression modeling on stratified data
• Covariate adjustment in stratified experiments
• Title not available (Why is that?)
• Rerandomization in Stratified Randomized Experiments
• Stratified doubly robust estimators for the average causal effect
• Toward Better Practice of Covariate Adjustment in Analyzing Randomized Clinical Trials
• Rejoinder on ‘Inference after covariate-adaptive randomization: aspects of methodology and theory’
• Randomization-based Joint Central Limit Theorem and Efficient Covariate Adjustment in Randomized Block 2 ^K Factorial Experiments

This page was built for publication: Regression-adjusted average treatment effect estimates in stratified randomized experiments

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5145706)