Better Experimental Design by Hybridizing Binary Matching with Imbalance Optimization

TL;DR

This paper introduces a hybrid experimental design method combining optimal nonbipartite matching with greedy switching to reduce covariate imbalance and improve treatment effect estimation robustness, especially for nonlinear response models.

Contribution

It proposes a novel hybrid design that enhances covariate balance and robustness by integrating matching with greedy switching heuristics, outperforming previous methods.

Findings

Significantly reduces covariate imbalance to rate $O_p(n^{-4})$

Improves mean squared error in nonlinear response models

Maintains robustness to unobserved covariates

Abstract

We present a new experimental design procedure that divides a set of experimental units into two groups in order to minimize error in estimating an additive treatment effect. One concern is minimizing error at the experimental design stage is large covariate imbalance between the two groups. Another concern is robustness of design to misspecification in response models. We address both concerns in our proposed design: we first place subjects into pairs using optimal nonbipartite matching, making our estimator robust to complicated non-linear response models. Our innovation is to keep the matched pairs extant, take differences of the covariate values within each matched pair and then we use the greedy switching heuristic of Krieger et al. (2019) or rerandomization on these differences. This latter step greatly reduce covariate imbalance to the rate $O_p(n^{-4})$ in the case of one covariate that are uniformly distributed. This rate benefits from the greedy switching heuristic which is $O_p(n^{-3})$ and the rate of matching which is $O_p(n^{-1})$. Further, our resultant designs are shown to be as random as matching which is robust to unobserved covariates. When compared to previous designs, our approach exhibits significant improvement in the mean squared error of the treatment effect estimator when the response model is nonlinear and performs at least as well when it the response model is linear. Our design procedure is found as a method in the open source R package available on CRAN called GreedyExperimentalDesign.