Doubly Robust Identification of Causal Effects of a Continuous Treatment using Discrete Instruments
报告人: Yingying Dong (University of California Irvine)
时间:2024-03-14 15:10-17:00
地点:Zoom (ID: 959 4068 3373 Passcode: 1234)
Abstract:
Many empirical applications estimate causal effects of a continuous endogenous variable (treatment) using a binary instrument. Estimation is typically done through linear 2SLS. This approach requires a mean treatment change and causal interpretation requires the LATE-type monotonicity in the first stage. An alternative approach is to explore distributional changes in the treatment, where the first-stage restriction is treatment rank similarity. We propose causal estimands that are doubly robust in that they are valid under either of these two restrictions. We apply the doubly robust estimation to estimate the impacts of sleep on well-being. Our new estimates corroborate the usual 2SLS estimates.
About the Speaker:
I am a professor of economics at UC Irvine. I am also a research fellow at IZA. My current research interests are causal identification and inference, treatment effect models and public policy evaluation. I am particularly interested in extending causal identification to be valid under more general conditions and finding novel solutions to identification problems that may not fit in the standard toolbox. A defining feature of my research is that most of my econometric work is grounded in prominent empirical issues, as I tend to concern myself with empirical phenomena first and then work toward finding econometric solutions. On the applied side, I have broad interests in education, labor and health topics.

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!