Finite-Sample Properties of Propensity-Score Matching and Weighting Estimators
- 1 February 2004
- journal article
- Published by MIT Press in The Review of Economics and Statistics
- Vol. 86 (1) , 77-90
- https://doi.org/10.1162/003465304323023697
Abstract
The finite-sample properties of matching and weighting estimators, often used for estimating average treatment effects, are analyzed. Potential and feasible precision gains relative to pair matching are examined. Local linear matching (with and without trimming), k-nearest-neighbor matching, and particularly the weighting estimators performed worst. Ridge matching, on the other hand, leads to an approximately 25% smaller MSE than does pair matching. In addition, ridge matching is least sensitive to the design choice. © 2004 President and Fellows of Harvard College and the Massachusetts Institute of Technology.Keywords
This publication has 15 references indexed in Scilit:
- Efficient Estimation of Average Treatment Effects Using the Estimated Propensity ScoreEconometrica, 2003
- The role of the propensity score in estimating dose-response functionsBiometrika, 2000
- Matching As An Econometric Evaluation EstimatorThe Review of Economic Studies, 1998
- On the Role of the Propensity Score in Efficient Semiparametric Estimation of Average Treatment EffectsEconometrica, 1998
- Matching As An Econometric Evaluation Estimator: Evidence from Evaluating a Job Training ProgrammeThe Review of Economic Studies, 1997
- Local Polynomial Regression: Optimal Kernels and Asymptotic Minimax EfficiencyAnnals of the Institute of Statistical Mathematics, 1997
- Local Regression: Automatic Kernel CarpentryStatistical Science, 1993
- Local Linear Regression Smoothers and Their Minimax EfficienciesThe Annals of Statistics, 1993
- Design-adaptive Nonparametric RegressionJournal of the American Statistical Association, 1992
- Variable Bandwidth and Local Linear Regression SmoothersThe Annals of Statistics, 1992