講者：陳宜廷教授（台大財金系教授／統計所合聘教師）。
時間：111年9月29日（星期四）13:30。
地點：台大次震宇宙館601室。
講題：Regularization of Synthetic Controls for Policy Evaluation
摘要：We propose a unified approach to compare various synthetic control (SC) methods and generate new SC methods. Our approach is built on a mean squared prediction error (MSPE) upper bound of an arbitrary SC method in predicting the counterfactual of a treated unit. This MSPE bound is established without assuming the true outcome model or imposing a combination restriction on the SC unit, and allows for the use of auxiliary models to refine the potential imperfect matching of SC. Our approach also includes a generalized SC method that regularizes the squared-bias and variance components of the MSPE bound. We show that the regularized SC method encompasses useful complements to existing SC methods by theoretical comparison, simulation and empirical illustration.