Doubly Robust Identification for Causal Panel Data Models

Dmitry Arkhangelsky, Guido W. Imbens

We study identification and estimation of causal effects in settings with panel data. Traditionally researchers follow model-based identification strategies relying on assumptions…