Standard Synthetic Control Methods: The Case of Using All Preintervention Outcomes Together With Covariates
研究发现，在标准合成控制方法中，若将全部干预前结果变量滞后项作为预测变量，其他协变量会变得无关紧要，即使它们对预测干预后结果很重要，这威胁估计的无偏性；限制滞后项使用可让其他协变量获得正权重，改变估计结果和政策结论。
It is becoming increasingly popular in applications of standard synthetic control methods to include the entire pretreatment path of the outcome variable as economic predictors. We demonstrate both theoretically and empirically that using all outcome lags as separate predictors renders all other covariates irrelevant in such settings. This finding holds irrespective of how important these covariates are for accurately predicting posttreatment values of the outcome, threatening the estimator’s unbiasedness. We show that estimation results and corresponding policy conclusions can change considerably when the usage of outcome lags as predictors is restricted, resulting in other covariates obtaining positive weights. Monte Carlo studies examine potential bias.