Session 1: A Course Correction in the History of Econometrics

The econometric workflow; treatment effects and binary choice; reduced form vs. structural; fixed vs. random design.

When: TBA · Where: TBA

Materials

• Slides (PDF)
• Recitation 1.1 slides (PDF): endogenous regressors in binary choice and the Heckman selection model
• Recording: to be posted after the session

Overview

• Why econometrics starts from a question, not an estimator
• Example I, vaccination and the ATE: specification, interference, identification by randomization
• Example II, binary choice: random utility, indifference, identification vs. optimizer convergence
• What “reduced form” means, and how to tell structural from reduced-form work in practice
• Fixed vs. random design: where the randomness comes from

Readings

• Angrist and Pischke (2009), Chapters 1–2.
• Heckman, J. J. (2010). “Building Bridges between Structural and Program Evaluation Approaches to Evaluating Policy.” Journal of Economic Literature, 48(2), 356–398.
• Low, H., and C. Meghir (2017). “The Use of Structural Models in Econometrics.” Journal of Economic Perspectives, 31(2), 33–58.