Difference-in-Differences (DiD) is among the most popular strategies to identify causal effects in observational studies. However, the literature is evolving fast, making it challenging to keep up with best practices. This is your chance to get up to speed! Enroll now to deepen your understanding of modern DiD tools and engage with one of the most accessible leading experts in the field, Pedro Sant’Anna. Not only will we cover the tools but also how to use them in practice!

Price: $995  $595

Student Price (also applicable for developing countries): $295 – contact us for the link

What this course is about

Difference-in-Differences (DiD) methods are widely used to answer what-if type of questions in economics, political science, and many other social and medical sciences. These methods are also very popular in industry, where causal inference play a prominent role.

Although very popular, the last few years have seen a booming of new papers on DiD and related designs, making it challenging to keep up with rapidly evolving best practices. The main goal of this course is to provide a fast-track towards these best practices, enabling each and every attendee to be comfortable with a wide range of DiD tools.

As we believe that the best way to really learn any data science tool is to blend its theory with real-life applications, each lecture session will include a hands-on exercise that illustrates the content covered.  We will provide practical guidance on implementing these tools in R and Stata!


We will cover the theory and practice of DiD methods in detail. We will practice the implementation of these tools using code and sample data.  The portions of the course which are marked as extra will be made available via recordings.

Topic 1: Introduction to DiD and a brief overview of causal inference

We will discuss DiD popularity by briefly highlighting several applications in different fields. We will also lay down the potential outcome framework that will serve as the foundation for all our discussions. More specifically, we will:

  • Highlight DiD popularity in different fields;
  • Introduce the potential outcome framework and how we use it to embrace treatment effect heterogeneity;
  • Discuss the challenges to conduct causal inference with observational data, and how DiD methods address these.

Topic 2: The Classical 2x2 DiD setup

We do a deep-dive into the classical two-periods and two-groups DiD setup, paying particular attention to the commonly used assumptions to justify its reliability. In this section of the course, we will discuss:

  • The role of identifying assumptions: no anticipation and parallel trends;
  • Estimation of average treatment effects using simple comparison of means and regressions;
  • Inference procedures and clustering;
  • Falsification tests;
  • When parallel trends assumption is sensitive to functional form restrictions (extra).

Topic 3: 2x2 DiD setups with covariates

In many situations researchers wish to leverage available information about observed characteristics in DiD setups. Here, we will describe how you can reliably do this. We will discuss the following topics:

  • Allowing for covariate-specific trends;
  • Pitfalls of some two-way fixed-effects linear regression specifications;
  • Estimating treatment effects using the
    • outcome-regression approach;
    • inverse probability weighting approach;
    • doubly robust approach;
  • How to use machine learning procedures to do DiD (extra);
  • Panel data vs. Repeated cross-sectional data (extra).

Topic 4: DiD with multiple time periods

It is often the case in applications that we have access to more than two time periods. In this section we will discuss how we can leverage the additional information coming from these extra time periods, in setups without variation in treatment timing. Our lectures will shed light on the following questions:

  • Do we need to change the parallel trends assumption?
  • Can we allow for treatment anticipation?
  • How can we characterize treatment effect dynamics?
  • Can we assess the validity of parallel trends assumptions?
  • What if parallel trends only hold “approximately’’?

Topic 5: DiD with variation in treatment timing

It is not uncommon to have units being exposed to treatment at different points in time. How do DiD procedures perform in these more challenging setups? Does the choice of estimation method matter? How so? The recent DiD literature has provided many insights in these cases, and we will cover a good chunk of that, particularly in the case with staggered treatment adoption. Our discussion topics include:

  • What are the causal parameters of interest?
  • What type of parallel trends are we willing to impose?
  • Pitfalls of two-way fixed effects linear regression specifications;
  • Recovering meaningful causal parameters;
  • Highlighting treatment effect dynamics via event-studies;
  • Highlighting other sources of treatment effect heterogeneity;
  • Falsification tests;
  • Triple-Differences (extra);
  • What if treatment timing is random? (extra)
    • Can we do better than DiD? (extra)
    • How can we do randomization tests? (extra)
    • Do this actually matter in practice? (extra)
  • Challenges with setups where treatment can turn on and off (extra).

About Your Instructor

Pedro H. C. Sant'Anna, Ph.D.

Pedro H. C. Sant’Anna, Ph.D., is an Associate Professor of Economics at Emory University and an Amazon Visiting Academic. Before joining Emory and Amazon, he was at Vanderbilt University and Microsoft.

Pedro is a passionate econometrician, working in causal inference and semi- and non-parametric methods (also known as machine learning). Much of his recent work aims to develop, better understand, and further improve Difference-in-Differences methods. In fact, Pedro is among the most impactful researchers working on Difference-in-Differences methods, with more than 5,000 citations on the topic. 

He has published in top journals in economics and has given guest lectures and seminars on DiD topics at leading universities around the world, including Harvard University, Yale University, University of Chicago, MIT, UC Berkeley, among others. He is also a coauthor of several open-source packages for Difference-in-Difference methods, allowing him to have a first-hand experience on practicalities of these modern tools. His industry experience also allows him to effectively communicate with a broad audience with many different backgrounds. 

Pedro received his BA in Economics from Ibmec-MG (Brazil) in 2009, his M.A. and Ph.D. in Economics at Universidad Carlos III de Madrid (Spain) in 2011 and 2015, respectively. He is a recipient of the 2016 Arnold Zellner Thesis Award in Econometrics and Statistics, awarded by the American Statistical Association.

For more information about Pedro, access his personal website: https://psantanna.com/ 

What people say about the course