Holder： Zhengyu Zhang（Shanghai University of Finance and Economics）
Time：2024-10-24 15:10-17:00
Location：Room 217, Guanghua Building 2
Abstract:
One of the main objectives of policy analysis is to evaluate the effect of a counterfactual change in a covariate D on the unconditional distribution of an outcome variable Y. The existing methods (Firpo et al. (2009), Martínez-Iriarte et al. (2024)) focus on the case when D is exogenous. When D is endogenous, identification of unconditional policy effect usually requires the use of instrumental variable. In practice, however, it may not be possible to find a valid instrument. In this talk, we show some unconditional policy effect estimands can be identified when no instrument is accessible.
We consider two cases. First, we consider identification and inference of a wide range of unconditional partial policy effects at the threshold in a regression kink (RK) design with general policy intervention. Our work is the first to discover and clarify the connections between many RK estimands in the prior literature and provide a unifying formula to understand the causal interpretation of some RK esitmands that are new in the literature.
Second, we consider identification of unconditional quantile partial effects in a nonseparable panel data model with time homogeneity. We find that unconditional effects are identified for the “stayers”, i.e. for individuals with the same regressor values in two time periods.
About the Speaker:
张征宇，上海财经大学经济学院数量经济系主任、教授、博士生导师。主要研究基于因果推断的中国经济政策评价。在《经济研究》、《管理世界》、 《经济学季刊》、Journal of Business and Economic Statistics, Econometric Theory 等期刊发表论文40多篇。先后主持国家自科、社科基金四项；还主持国家自科重点项目子课题与国家社科重大项目子课题。2023年获国际计量经济学权威期刊 Econometric Theory 颁发的 Multa Scripsit Award；以第一完成人获第九届高等学校科学研究优秀成果三等奖；上海市哲学社会科学优秀成果一等奖、二等奖。入选教育部经济学“101计划”《计量经济学》课程编委会。出版专著《异质性政策效应的识别和推断：理论与中国案例》与教材《计量经济简明教程》（高等教育出版社）；担任期刊《系统工程理论与实践》客座主编。
Your participation is warmly welcomed!
欢迎扫码关注北大统计科学中心公众号，了解更多讲座信息!
