Revisiting regression adjustment in experiments with heterogeneous treatment effects

By Akanksha Negi, Jeffrey M. Wooldridge. 2020

74 citations recorded by the source corpus.

Abstract In the context of random sampling, we show that linear full (separate) regression adjustment (FRA) on the control and treatment groups is, asymptotically, no less efficient than both the simple difference-in-means estimator and the pooled regression adjustment estimator; with heterogeneous treatment effects, FRA is usually strictly more efficient. We also propose a class of nonlinear regression adjustment estimators where consistency is ensured despite arbitrary misspecification of the conditional mean function. A simulation study confirms that nontrivial efficiency gains are possible with linear FRA, and that further gains are possible, even under severe mean misspecification, using nonlinear FRA.

Subfields: Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/4367bb382fe08e4f9be801134847d5efb758904e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Efficient Estimation for Staggered Rollout Designs (2021, background): “Negi and Wooldridge (2021) study covariate adjustment in cross-sectional randomized experiments from a superpopulation perspective and show that Lin ’ s (2013) estimator (which they refer to as full regression adjustment [FRA]) achieves the semiparametric efﬁciency bound when the conditional…”