STA 290 Seminar: Sylvia Klosin

Abstract: This talk presents work on causal inference in panel data motivated by environmental and climate applications. I discuss three projects: one on dynamic bias in fixed effects estimators, one on machine learning methods for continuous treatments in panel data, and one on design-based methods and randomization inference for persistent, path-dependent treatments. Across these projects, the goal is to develop methods that allow for richer dynamics and more flexible treatment settings while remaining useful for applied researchers.

 

Bio: Sylvia Klosin is an Assistant Professor in Agricultural and Resource Economics at UC Davis. She received her PhD in Economics from MIT and was also part of MIT’s Interdisciplinary PhD in Statistics program. Her research lies at the intersection of econometrics and environmental economics. She develops causal inference methods for panel data, with interests in machine learning, high-dimensional statistics, and dynamic treatment effects. Her work is motivated by applications in agricultural and environmental economics.