View in EDS

Optimal rerandomization designs via a criterion that provides insurance against failed experiments.

Saved in:
Bibliographic Details
Title: Optimal rerandomization designs via a criterion that provides insurance against failed experiments.
Authors: Kapelner, Adam1 (AUTHOR) kapelner@qc.cuny.edu, Krieger, Abba M.2 (AUTHOR), Sklar, Michael3 (AUTHOR), Azriel, David4 (AUTHOR)
Source: Journal of Statistical Planning & Inference. Jul2022, Vol. 219, p63-84. 22p.
Subjects: Confidence regions (Mathematics), Confidence intervals, Randomization (Statistics), Insurance, Experimental design, Treatment effectiveness
Abstract: We present an optimized rerandomization design procedure for a non-sequential treatment-control experiment. Randomized experiments are the gold standard for finding causal effects in nature. But sometimes random assignments result in unequal partitions of the treatment and control group visibly seen as imbalance in observed covariates. There can additionally be imbalance on unobserved covariates. Imbalance in either observed or unobserved covariates increases treatment effect estimator error inflating the width of confidence regions and reducing experimental power. "Rerandomization" is a strategy that omits poor imbalance assignments by limiting imbalance in the observed covariates to a prespecified threshold. However, limiting this threshold too much can increase the risk of contracting error from unobserved covariates. We introduce a criterion that combines observed imbalance while factoring in the risk of inadvertently imbalancing unobserved covariates. We then use this criterion to locate the optimal rerandomization threshold based on the practitioner's level of desired insurance against high estimator error. We demonstrate the gains of our designs in simulation and in a dataset from a large randomized experiment in education. We provide an open source R package available on CRAN named OptimalRerandExpDesigns which generates designs according to our algorithm. • The rerandomization experimental design threshold can be optimized. • Our optimization considers both the effects of observed and unobserved covariates. • Our optimization improves upon linear regression for general response models. • We provide a CRAN package that implements three optimization procedures. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Statistical Planning & Inference is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:We present an optimized rerandomization design procedure for a non-sequential treatment-control experiment. Randomized experiments are the gold standard for finding causal effects in nature. But sometimes random assignments result in unequal partitions of the treatment and control group visibly seen as imbalance in observed covariates. There can additionally be imbalance on unobserved covariates. Imbalance in either observed or unobserved covariates increases treatment effect estimator error inflating the width of confidence regions and reducing experimental power. "Rerandomization" is a strategy that omits poor imbalance assignments by limiting imbalance in the observed covariates to a prespecified threshold. However, limiting this threshold too much can increase the risk of contracting error from unobserved covariates. We introduce a criterion that combines observed imbalance while factoring in the risk of inadvertently imbalancing unobserved covariates. We then use this criterion to locate the optimal rerandomization threshold based on the practitioner's level of desired insurance against high estimator error. We demonstrate the gains of our designs in simulation and in a dataset from a large randomized experiment in education. We provide an open source R package available on CRAN named OptimalRerandExpDesigns which generates designs according to our algorithm. • The rerandomization experimental design threshold can be optimized. • Our optimization considers both the effects of observed and unobserved covariates. • Our optimization improves upon linear regression for general response models. • We provide a CRAN package that implements three optimization procedures. [ABSTRACT FROM AUTHOR]
ISSN:03783758
DOI:10.1016/j.jspi.2021.11.005