方法对比

	机器学习增强匹配估计器	倾向得分匹配
领域	因果推断	研究统计学
方法族	Regression model	Process / pipeline
起源年份	2006–2018	1983
提出者	Abadie & Imbens (classical matching); Chernozhukov et al. (ML augmentation framework)	Paul Rosenbaum and Donald Rubin
类型	Causal inference / nonparametric matching	Method
开创性文献	Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21(1), C1-C68.	Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55.
别名	ML-augmented matching, ML matching estimator, high-dimensional matching estimator, data-adaptive matching estimator	PSM, propensity score weighting, covariate balance
相关	5	3
摘要	The machine learning-augmented matching estimator combines classical nearest-neighbor or propensity-score matching with ML algorithms — such as lasso, random forests, or gradient boosting — to select covariates, estimate propensity scores, and correct for residual bias. The result is a matching-based causal estimator that remains valid under high-dimensional confounding where traditional hand-specified matching fails.	Propensity score matching (PSM) is a method for reducing confounding bias in observational studies by balancing baseline characteristics between treatment groups, simulating randomization. Developed by Rosenbaum and Rubin (1983), it estimates the probability of receiving treatment given observed covariates, then matches or weights treated and control individuals with similar treatment probabilities. Widely used in medicine, epidemiology, and policy evaluation when randomized trials are infeasible or unethical, enabling estimation of treatment effects while controlling for selection bias.