ECONOMETRICS

Scientific publications in management science are increasingly relying on program evaluation techniques, which therefore need to be part of the standard training of graduate management students.

The course will endow the students with the basic tools to deal with causal questions. The students will familiarize with the most used program evaluation techniques and more in general will be able to address the identification and estimation issues arising in the quasi- experimental and non-experimental context.

Maths, probability, statistics.

1. The experimental ideal and the selection problem, [AP] pp. 11-24.
2. The anatomy of linear regression, [AP] pp. 27-48.
3. Using the linear model to describe causal relationship, [AP] pp. 51-68.
4. Regression meets matching, [AP] pp. 69-77.
5. The fundamental role of propensity score in observational studies, [AP]
pp. 80-91.
6. Instrumental variables and causality, [AP] pp. 113-121.
7 . Two-stage least squares, [AP] pp. 121-127, [AP] pp. 138-147, [AP] pp.
190-192, [AP] pp. 205-216.
8. The Wald Estimator, [AP] pp. 127-133.
9. Instrumental variables with heterogeneous potential outcome, [AP] pp.
150-172.
10. Fixed effects, [AP] pp. 221-227 and [SS].
11. Diff-in-diff and event studies design, [AP] pp. 227-243.
12. RD design, [AP] pp. 251-267.

[AP] Angrist J.D. and S. Pischke, Mostly Harmless Econometrics: An Empiricist’s Companion, 2009, Princeton University Press.
de Chaisemartin, C., & D'Haultfœuille, X. (2020). Two-way fixed effects estimators with heterogeneous treatment effects. American Economic Review, 110, 2964–2996.
Schmidheiny, K. and S.Siegloch, On event studies and distributed-lags in two-way fixed effects models: Identification, equivalence, and generalization, Journal of Applied Econometrics, in press.
Wooldridge, J.M. , Econometric Analysis of Cross Section and Panel Data, 2002, The MIT press.

The exam is made of two components the take-home and in-class exams. Both exams are written. The take-home exam is a computer coding exercise. Students are expected to be able to carry with real data a causal effects estimation exercises. The in-class exam is made of True/False questions and exercises. The grade is 18 put of 32 in each of the two components and the exam is passed if the minimum between the grades on each of the two component is above 18. The final grade is the arithmetic mean of the grades on the take-home and in-class exam.

In person lecturing

written