Abstract:
|
Matching estimators for average treatment effects are widely used in the binary treatment setting, in which missing potential outcomes are imputed as the average of observed outcomes of all matches for each unit. With more than two treatment groups, however, estimation using matching requires additional assumptions and techniques. In this paper, we propose a nearest-neighbor matching estimator for use with multiple, nominal treatments, and use an extensive simulation with vector matching to show that this method is efficient and has coverage levels that are close to nominal. We conclude by illustrating limitations in both the matching and estimation procedures, and discuss several areas for future research. These results provide new conclusions about estimating causal effects in the multiple treatment setting.
|