More Efficient Policy Learning via Optimal Retargeting
Author:
- 3 August 2020
- journal article
- research article
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 116 (534) , 646-658
Abstract
Policy learning can be used to extract individualized treatment regimes from observational data in healthcare, civics, e-commerce, and beyond. One big hurdle to policy learning is a commonplace lack of overlap in the data for different actions, which can lead to unwieldy policy evaluation and poorly performing learned policies. We study a solution to this problem based on retargeting, that is, changing the population on which policies are optimized. We first argue that at the population level, retargeting may induce little to no bias. We then characterize the optimal reference policy and retargeting weights in both binary-action and multi-action settings. We do this in terms of the asymptotic efficient estimation variance of the new learning objective. We further consider weights that additionally control for potential bias due to retargeting. Extensive empirical results in a simulation study and a case study of personalized job counseling demonstrate that retargeting is a fairly easy way to significantly improve any policy learning procedure applied to observational data.Keywords
All Related Versions
This publication has 30 references indexed in Scilit:
- Private and Public Provision of Counseling to Job Seekers: Evidence from a Large Controlled ExperimentAmerican Economic Journal: Applied Economics, 2014
- Estimating Individualized Treatment Rules Using Outcome Weighted LearningJournal of the American Statistical Association, 2012
- Multivariate Matching Methods That Are Monotonic Imbalance BoundingJournal of the American Statistical Association, 2011
- Minimax regret treatment choice with finite samplesJournal of Econometrics, 2009
- Truncated Importance SamplingJournal of Computational and Graphical Statistics, 2008
- Does matching overcome LaLonde's critique of nonexperimental estimators?Journal of Econometrics, 2005
- Efficient Estimation of Average Treatment Effects Using the Estimated Propensity ScoreEconometrica, 2003
- Causal Effects in Nonexperimental Studies: Reevaluating the Evaluation of Training ProgramsJournal of the American Statistical Association, 1999
- On the Role of the Propensity Score in Efficient Semiparametric Estimation of Average Treatment EffectsEconometrica, 1998
- Estimation of Regression Coefficients When Some Regressors are not Always ObservedJournal of the American Statistical Association, 1994