Last updated: July 2026.

This post provides an up-to-date overview of Difference-In-Difference tools, papers, and other links. What is important here is not the timeliness but the relevance. While there are new methods daily, only a few are used in practice (or demanded by reviewers). The idea is that not only the advantages of each method are given here, but also the disadvantages. Unfortunately, disadvantages of individual methods (e.g. scalability to larger data sets) are rarely communicated.

Packages

Most of these tools were written by econometricians working in R, and that is still where coverage is widest. What has changed since I first wrote this post is that Python and Julia are no longer an afterthought: pyfixest now ports much of the fixest interface, including the staggered-adoption estimators, and Julia’s FixedEffectModels.jl is fast enough to matter on large panels. The replication package for our own paper runs in all three. R is still the safest default if you want a method the day it is published, but it is no longer the only option.

Core estimators

  • did: the Callaway & Sant’Anna estimator. The reference implementation. fastdid is a drop-in alternative when the panel gets large.
  • fixest: very fast fixed effects, with Sun & Abraham (2021) available out of the box. The workhorse for most of what I do.
  • did2s: Gardner’s two-stage estimator. Cheap to run and easy to explain in a paper.
  • did_imputation: the Borusyak, Jaravel & Spiess imputation estimator, efficient under homoskedasticity.

Sensitivity and diagnostics

  • HonestDiD: sensitivity analysis for violations of parallel trends. Increasingly requested by reviewers, so worth running before submission rather than after.

Synthetic control

  • synthdid: synthetic difference-in-differences.
  • tidysynth: the classical synthetic control method with a readable interface.

Weighting and matching

  • WeightIt: an alternative to PSM that weights observations instead of filtering them.
  • cobalt: balance diagnostics for whatever you used to build the comparison group.

Time series

  • CausalImpact: Bayesian structural time series, useful when you have one treated unit and a long pre-period.

Overviews & Tutorials

While online tutorials are numerous, I will link those that helped me here:

Sub-Topics

FAQ

  • Is it easy to add controls to a DiD? Not always
  • Is it appropriate to use a log-transformed DV in a DiD setting? It is often done, but there are some issues
  • What if my outcome has zeros, or a very heavy tail? Then the choice of transformation is not cosmetic: it silently changes which estimand you are targeting. See Chen & Roth (2024) below, and the DiD Estimand Lab for a hands-on version of the problem.

Papers

Here are listed those papers that you should know, grouped by the problem they solve.

Staggered adoption and event studies

  • Goodman-Bacon, A. (2021). Difference-in-differences with variation in treatment timing. Journal of Econometrics, 225(2), 254-277. DOI
  • de Chaisemartin, C., & D’Haultfœuille, X. (2020). Two-way fixed effects estimators with heterogeneous treatment effects. American Economic Review, 110(9), 2964-2996. DOI
  • Callaway, B., & Sant’Anna, P. H. (2021). Difference-in-differences with multiple time periods. Journal of Econometrics, 225(2), 200-230. PDF
  • Sun, L., & Abraham, S. (2021). Estimating dynamic treatment effects in event studies with heterogeneous treatment effects. Journal of Econometrics, 225(2), 175-199. PDF
  • Borusyak, K., Jaravel, X., & Spiess, J. (2024). Revisiting event-study designs: Robust and efficient estimation. Review of Economic Studies, 91(6), 3253-3285. DOI
  • de Chaisemartin, C., & d’Haultfoeuille, X. (2023). Two-way fixed effects and differences-in-differences with heterogeneous treatment effects: A survey. The Econometrics Journal, 26(3), C1-C30. PDF

Parallel trends, functional form, and heavy tails

  • Roth, J., & Sant’Anna, P. H. (2023). When is parallel trends sensitive to functional form? Econometrica, 91(2), 737-747. DOI
  • Rambachan, A., & Roth, J. (2023). A more credible approach to parallel trends. Review of Economic Studies, 90(5), 2555-2591. PDF
  • Chen, J., & Roth, J. (2024). Logs with zeros? Some problems and solutions. The Quarterly Journal of Economics, 139(2), 891-936. DOI
  • Wooldridge, J. M. (2023). Simple approaches to nonlinear difference-in-differences with panel data. The Econometrics Journal, 26(3), C31-C66. DOI
  • Ciani, E., & Fisher, P. (2019). Dif-in-dif estimators of multiplicative treatment effects. Journal of Econometric Methods, 8(1). DOI
  • Santos Silva, J. M. C., & Tenreyro, S. (2006). The log of gravity. The Review of Economics and Statistics, 88(4), 641-658. DOI
  • Winkler, D., Hotz-Behofsits, C., Wlömert, N., Papies, D., & Liaukonytė, J. (2026). Does TikTok promote or cannibalize music streaming? Estimands and identification with heavy-tailed outcomes. Quantitative Marketing and Economics (forthcoming). SSRN, practitioner’s companion

Synthetic control and time series

  • Abadie, A. (2021). Using synthetic controls: Feasibility, data requirements, and methodological aspects. Journal of Economic Literature, 59(2), 391-425. DOI
  • Arkhangelsky, D., Athey, S., Hirshberg, D. A., Imbens, G. W., & Wager, S. (2021). Synthetic difference-in-differences. American Economic Review, 111(12), 4088-4118. DOI
  • Brodersen, K. H., Gallusser, F., Koehler, J., Remy, N., & Scott, S. L. (2015). Inferring causal impact using Bayesian structural time-series models. The Annals of Applied Statistics, 247-274. PDF

Surveys and adjacent methods

  • Roth, J., Sant’Anna, P. H., Bilinski, A., & Poe, J. (2023). What’s trending in difference-in-differences? A synthesis of the recent econometrics literature. Journal of Econometrics. PDF
  • Goldfarb, A., Tucker, C., & Wang, Y. (2022). Conducting research in marketing with quasi-experiments. Journal of Marketing, 86(3), 1-20. DOI
  • Athey, S., & Imbens, G. W. (2017). The state of applied econometrics: Causality and policy evaluation. Journal of Economic Perspectives, 31(2), 3-32. PDF
  • Wager, S., & Athey, S. (2018). Estimation and inference of heterogeneous treatment effects using random forests. Journal of the American Statistical Association, 113(523), 1228-1242. PDF

Books

  • Pearl, J. (2009). Causality. Cambridge university press.
  • Cunningham, S. (2021). Causal inference: The mixtape. Yale university press.
  • Angrist, J. D., & Pischke, J. S. (2009). Mostly harmless econometrics: An empiricist’s companion. Princeton university press.