Building Representative Matched Samples With Multi-Valued Treatments in Large Observational Studies

In this article, we present a new way of matching in observational studies that overcomes three limitations of existing matching approaches. First, it directly balances covariates with multi-valued treatments without explicitly estimating the generalized propensity score. Second, it builds self-weighted matched samples that are representative of a target population by design. Third, it can handle large datasets, with hundreds of thousands of observations, in a couple of minutes. The key insights of this new approach to matching are balancing the treatment groups relative to a target population and positing a linear-sized mixed integer formulation of the matching problem. We formally show that this formulation is more effective than alternative quadratic-sized formulations, as its reduction in size does not affect its strength from the standpoint of its linear programming relaxation. We also show that this formulation can be used for matching with distributional covariate balance in polynomial time under certain assumptions on the covariates and that it can handle large datasets in practice even when the assumptions are not satisfied. This algorithmic characterization is key to handling large datasets. We illustrate this new approach to matching in both a simulation study and an observational study of the impact of an earthquake on educational attainment. With this approach, the results after matching can be visualized with simple and transparent graphical displays: while increasing levels of exposure to the earthquake have a negative impact on school attendance, there is no effect on college admission test scores. Supplementary materials for this article are available online.