On regression adjustments in experiments with several treatments

Abstract: Regression adjustments are often made to experimental data. Since randomization does not justify the models, bias is likely; nor are the usual variance calculations to be trusted. Here, we evaluate regression adjustments using Neyman's nonparametric model. Previous results are generalized, and more intuitive proofs are given. A bias term is isolated, and conditions are given for unbiased estimation in finite samples.

Recommendations

Cites work

Cited in

(40)

Regression-based multiple treatment effect estimation under covariate-adaptive randomization
scientific article; zbMATH DE number 5575449 (Why is no real title available?)
A Design-Based Perspective on Synthetic Control Methods
Covariate adjustment in randomized clinical trials with missing covariate and outcome data
Rematching on-the-fly: sequential matched randomization and a case for covariate-adjusted randomization
Regression-adjusted estimation of quantile treatment effects under covariate-adaptive randomizations
Regression-assisted inference for the average treatment effect in paired experiments
Statistical inference of heterogeneous treatment effect based on single-index model
Bias of the regression estimator for experiments using clustered random assignment
On equivalencies between design-based and regression-based variance estimators for randomized experiments
Covariance adjustment in radomized experiments and observational studies
Sampling‐based Randomised Designs for Causal Inference under the Potential Outcomes Framework
Randomization does not justify logistic regression
A unified analysis of regression adjustment in randomized experiments
Inverse Probability Weighting Estimation in Completely Randomized Experiments
Regression adjustment for treatment effect with multicollinearity in high dimensions
Regression adjustment in completely randomized experiments with a diverging number of covariates
More powerful multiple testing in randomized experiments with non-compliance
Sharp bounds on the variance in randomized experiments
Regression-adjusted average treatment effect estimates in stratified randomized experiments
From Local to Global: External Validity in a Fertility Natural Experiment
Revisiting regression adjustment in experiments with heterogeneous treatment effects
On regression adjustments to experimental data
scientific article; zbMATH DE number 3860212 (Why is no real title available?)
The Perils of Balance Testing in Experimental Design: Messy Analyses of Clean Data
Leveraging population outcomes to improve the generalization of experimental results: application to the JTPA study
Randomization-based test for censored outcomes: a new look at the logrank test
Randomization-based causal inference from split-plot designs
Lasso adjustments of treatment effect estimates in randomized experiments
Toward Better Practice of Covariate Adjustment in Analyzing Randomized Clinical Trials
Agnostic notes on regression adjustments to experimental data: reexamining Freedman's critique
Propensity score weighting for covariate adjustment in randomized clinical trials
Using standard tools from finite population sampling to improve causal inference for complex experiments
Inference after covariate-adaptive randomisation: aspects of methodology and theory
Regression analysis of experiments with complex confounding patterns guided by the alias matrix.
The Generalized Oaxaca-Blinder Estimator
Rejoinder on ‘Inference after covariate-adaptive randomization: aspects of methodology and theory’
Covariate adjustment in randomization-based causal inference for \(2^K\) factorial designs
Front-door versus back-door adjustment with unmeasured confounding: bias formulas for front-door and hybrid adjustments with application to a job training program
A causal bootstrap