主题：Estimating ConditionalAverage Treatment Effects
主讲人：台湾中央研究院助理研究员 许育进
主持人：西南财经大学 郭萌萌副教授
时间：2014年3月14日下午2：30－4点
地点：柳林校区颐德楼H513
主办单位：经济与管理研究院 科研处
主讲人简介：
许育进，2010年毕业于美国德州大学奥斯汀分校经济系，现为台湾中央研究院助理研究员。他的主要研究方向为计量经济学理论。他在Journal of Econometrics, Journalof Business and Economic Statistics, Econometrics Journal,Economics Letters, Journal of Financial Econometrics, Journal of EmpiricalFinance, 等英文期刊上发表论文多篇。
内容提要：
This paper considers afunctional parameter called the conditional average treatment effect (CATE),designed to capture heterogeneity of a treatment effect across subpopulationswhen the unconfoundedness assumption applies. In contrast to quantileregressions, the subpopulations of interest are defined in terms of thepossible values of a set of continuous covariates rather than the quantiles ofthe potential outcome distributions. We show that the CATE parameter isnonparametrically identified under the unconfoundedness assumption and proposeinverse probability weighted estimators for it. Under regularity conditions,some of which are standard and some of which are new in the literature, we show(pointwise) consistency and asymptotic normality of a fully nonparametric and asemiparametric estimator. We apply our methods to estimate the average effectof a first-time mother's smoking during pregnancy on the baby's birth weight asa function of per capita income in the mother's zip code. For non-whitemothers, the average effect of smoking is predicted to become stronger (morenegative) as a function of income.
