Mostly Harmless Econometrics

A reading group on applied econometrics

About

This reading group works through Mostly Harmless Econometrics (Angrist and Pischke, 2009) and related readings, organized around one idea:

An estimator is not a research design.

Every session follows the same sequence of questions:

Question → Specification & Estimand → Identification → Estimation → Inference

The methods change from session to session. The sequence of questions does not.

Logistics

When TBA
Where TBA
Organizer Kaiwen Luo

Tentative topics

  1. A course correction in the history of econometrics
  2. Regression and selection on observables
  3. Instrumental variables and LATE
  4. Difference-in-differences
  5. Regression discontinuity
  6. Panel data
  7. Treatment-effect heterogeneity
  8. Structural models and policy counterfactuals

Slides, readings, and recordings for each meeting are on the Sessions page.

Core readings

  • Angrist, J. D., and J.-S. Pischke (2009). Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton University Press.
  • Lewbel, A. (2019). “The Identification Zoo: Meanings of Identification in Econometrics.” Journal of Economic Literature, 57(4), 835–903. doi:10.1257/jel.20181361