New Developments in Difference-in-Differences Estimators

About

The ISEG Winter School 2026 takes place from January 19-21, 2026 and is part of a course aimed at presenting and discussing the most recent theoretical and empirical advances in the scientific areas of applied economics and methods, and is organized by ISEG in partnership with the LESE

Course leader

Xavier D'Haultfoeuille, ENSAE Paris, Polytechnic Institute of Paris

Target group

This training is aimed at Masters, PhD students, researchers and university professorsand others Executive Management in the public administration, state business and private sectors

Course aim

Difference-in-differences are one of the most popular methods to estimate causal effects. But what do we mean exactly by "Difference-in-differences", once we depart from the original setup where some groups become to receive a binary treatment, whereas some others do not? And are there alternatives to the usual parallel-trends assumption? This course will review recent work on difference-in-differences that address these issues. It will follow the structure of our textbook in progress with Clément de Chaisemartin. The duration given for each section should be seen as indicative.

Introduction and set-up (1 hour)
Classical designs (2 hours)
Alternatives to parallel trends (2 hours)
TWFE estimators outside of the classical design (1 hour)
Designs with variation in treatment timing (2 hours)
Designs with variation in treatment dose (2 hours)
General designs (2 hours)

Fee info