Dynamic Weighting for Synthetic Control Method

洪少新，山东大学经济研究院助理研究员，主要从事计量经济学和统计学的理论和应用研究工作，具体研究方向包括非平稳时间序列分析、模型平均和高维数据统计推断等。主要工作发表于Econometrics Journal、Economics Letters和Journal of Financial Econometrics等。主持国家自然科学基金青年项目。

We propose a novel synthetic control method with a dynamic weighting scheme to evaluate the impacts of social policy. The basic idea is to utilize the interdependence between different control units in a panel dataset to create the counterfactuals locally. Unlike the existing literature, we allow the weights to change over state variables and thus it is expected to capture potentially nonlinear features in economics and finance. It is shown that the treatment-effect estimator is asymptotically optimal in the sense of achieving the lowest possible local squared prediction error. The rate of the selected weights converging to the optimal weights to minimizing the expected local quadratic loss is established. Simulations and empirical applications are conducted to evaluate the finite sample performance of the proposed method.