Better experimental design by hybridizing binary matching with imbalance optimization

DOI10.1002/CJS.11685MaRDI QIDQ6059445zbMATH OpenOpenAlexFDO

Authors Abba M. Krieger, David Azriel, Adam Kapelner

Publication date 2 November 2023

Published in The Canadian Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2012.03330

zbMATH Keywords

experimental designmatchingoptimizationrerandomization

Mathematics Subject Classification ID

Statistics (62-XX)

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.

Cites work

• A matching procedure for sequential experiments that iteratively learns which covariates improve power
• Agnostic notes on regression adjustments to experimental data: reexamining Freedman's critique
• Asymptotic theory of rerandomization in treatment-control experiments
• Bias-Corrected Matching Estimators for Average Treatment Effects
• Forcing a sequential experiment to be balanced
• Harmonizing Optimized Designs With Classic Randomization in Experiments
• scientific article; zbMATH DE number 1220667 (Why is no real title available?)
• Matching methods for causal inference: a review and a look forward
• Matching on-the-fly: sequential allocation with higher power and efficiency
• Nearly random designs with greatly improved balance
• Observational studies.
• Optimal multivariate matching before randomization
• OptimalA PrioriBalance in the Design of Controlled Experiments
• Probability for statistics and machine learning. Fundamentals and advanced topics.
• Randomization in clinical trials. Theory and practice
• Rerandomization Strategies for Balancing Covariates Using Pre-Experimental Longitudinal Data
• Rerandomization to balance tiers of covariates
• Rerandomization to improve covariate balance in experiments
• Ridge rerandomization: an experimental design strategy in the presence of covariate collinearity
• Sequential rerandomization
• The power of optimization over randomization in designing experiments involving small samples
• Using Multivariate Matched Sampling and Regression Adjustment to Control Bias in Observational Studies

Cited in

(3)

• GreedyExperimentalDesign
• The role of pairwise matching in experimental design for an incidence outcome
• Asymptotic theory of the best-choice rerandomization using the Mahalanobis distance

This page was built for publication: Better experimental design by hybridizing binary matching with imbalance optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6059445)