External Seminar by Sukjin Han (University of Bristol)

Title: Semiparametric Models for Dynamic Treatment Effects and Mediation Analyses with Observational Data

Abstract: This paper proposes a semiparametric model that captures how a sequence of interventions interacts with a sequence of outcomes. In this setup, the outcome at the given period is affected by the history of treatments and outcomes, directly or indirectly through mediators. The main challenge in understanding various channels of dynamic effects is that, in observational settings, individuals make dynamically endogenous decisions whether to select into treatments. Using the approach of instrumental variables, this paper shows how the average and quantile dynamic treatment effects and mediation effects can be point identified and efficiently estimated in a class of semiparametric models under treatment endogeneity and flexible heterogeneity. Our procedure only requires binary instruments. As a byproduct of our semiparametric specification, we also identify and estimate parameters that reflect the degree of endogenous selection and time-invariant heterogeneity.

You can find out more about Sukjin’s research here: https://sukjinhan.com/