Diff-in-Diff in Panel Data

Overview

This is our third module on difference-in-differences estimation. We discuss a range of approaches to testing the common trends assumption, and introduce the event study framework for estimating treatment effects that evolve over time.

This module contains 38 minutes of video lecture. There are currently no readings or empirical exercises for this module.

Video Lectures

Lecture 6.1: Testing Common Trends (16:46)

Review Questions

1. What is the common trends assumption, and why is it important in difference-in-differences estimation?
2. To what extent can the common trends assumption be tested, and under what circumstances?
3. What are the three approaches to testing common trends? Provide an example of each?

Lecture 6.2: Event Studies (20:52)

Review Questions

1. What is the event study approach, and how can it be used to evaluate the impacts of a program?
2. When might it make sense to use an event study approach rather than a simple 2X2 differences-in-differences estimation strategy?
3. How would you implement the event study approach?
4. How can you use the event study approach to test the common trends assumption?
5. What other hypothesis tests might you run after an event study? How might they be implemented?

Empirical Exercise

The empirical exercise for this module is not publicly available (as of March 2021). We hope to make the exercise available to the public in the future.