Lightnews — Scholar-powered news

Clément de Chaisemartin @cdechaisemartin.bsky.social · Jun 25

4/4

📖 Draft available for free until publication:
👉 papers.ssrn.com/sol3/papers....

We welcome feedback, praise, bug reports, and citations 😇
#DID #causalinference #econometrics #socialscience #bookdrop

Credible Answers to Hard Questions: Differences-in-Differences for Natural Experiments

<div> The purpose of this book is to introduce applied researchers to </div> <div> modern Differences-in-Differences (DID) estimators, tailored to potentially

papers.ssrn.com

1 21

Clément de Chaisemartin @cdechaisemartin.bsky.social · Jun 25

3/4

Real examples, real confusion, real fixes.

We’ve been asked many many many hard DID questions over the years.
This book tries to answer them — with proofs & packages.

Bonus: A practitioners' checklist.

1 2

Clément de Chaisemartin @cdechaisemartin.bsky.social · Jun 25

2/4

What’s inside?

✅ A survival guide for DID in the wild
📉 When TWFE goes rogue
⚠️ When trends aren’t so parallel
🔍 When your control group is... questionable

Methods illustrated by revisiting 4 empirical articles so yes, there's code (Stata. R coming soon. Python fans, courage.).

1 1 4

Clément de Chaisemartin @cdechaisemartin.bsky.social · Jun 25

🧵1/4

🚨 We have finalized our DID textbook!
You’ve run out of excuses for sketchy pre-trends and mysterious TWFE coefficients.

📘 Credible Answers to Hard Questions: Differences-in-Differences for Natural Experiments
By Chaisemartin & D’Haultfœuille.

papers.ssrn.com/sol3/papers....

Credible Answers to Hard Questions: Differences-in-Differences for Natural Experiments

<div> The purpose of this book is to introduce applied researchers to </div> <div> modern Differences-in-Differences (DID) estimators, tailored to potentially

papers.ssrn.com

2 64 230

Clément de Chaisemartin @cdechaisemartin.bsky.social · Jun 23

but governments—who fund the research behind these tools—apply them too rarely to their own policies. In this piece, Xavier D'Haultfoeuille and I argue for greater use of evaluation tools in public decision-making and for ensuring that these tools remain accessible public goods.

4

Clément de Chaisemartin @cdechaisemartin.bsky.social · Jun 23

Pleased to share op-ed in Le Monde: "Impact Evaluation: For GAFAM or for Citizens?"

🔗 French version (Le Monde):
www.lemonde.fr/idees/articl...

🔗 English translation (Sciences Po):
www.sciencespo.fr/department-e...

Tech companies routinely use impact evaluation tools to optimize their services...

« Les outils d’évaluation sont sous-utilisés par ceux qui les financent, à savoir les Etats »

TRIBUNE. L’évaluation d’impact, qui permet par exemple de tester l’efficacité des politiques publiques, est massivement captée par les géants du numérique, alors que ces outils sont souvent créés par ...

www.lemonde.fr

1 4 11

Reposted by Clément de Chaisemartin

Randy Besco @randybesco.bsky.social · Jun 3

Looking forward to presenting “Electoral Discrimination Reconsidered”, with Selen Onculoglu at #CPSA_ACSP25

The last few years have seen a) major changes in racial politics b) new null results c) methodological developments.

So: do ethnic minority candidates really get less votes?

1 2 4

Clément de Chaisemartin @cdechaisemartin.bsky.social · May 21

Very excited to teach, with Moussa Richard Kouamé, in the 2025 Africa Summer School of the Econometrics Society, in ENSEA Abidjan! The class will be about DIDs (what else?), following our textbook. Looking forward to it! www.econometricsociety.org/regional-act...
papers.ssrn.com/sol3/papers....

2025 Africa Summer School - The Econometric Society

We are happy to announce that the 2025 Africa Summer School will take place in Abidjan,...

www.econometricsociety.org

1 5

Reposted by Clément de Chaisemartin

Carlisle Rainey 👨‍💻📊📚 @carlislerainey.bsky.social · May 21

new preprint

"On the Foundations of the Design-Based Approach"

from Aronow, Jang, and Offer-Westort

1 3 10

Reposted by Clément de Chaisemartin

Cannon "like boom" Cloud @cannoncloud.bsky.social · May 7

I created a shiny web tool to play around with OL/TWFE so you can teach how this stuff can get so screwy. You can mess with temporal and cohort heterogeneity, treatment timing, whether you have any controls, etc. Share your worst plots! #econsky cannoncloud.shinyapps.io/TWFE_OLS_Pla...

TWFE Event_study plot showing sign reversal

5 19 58

Clément de Chaisemartin @cdechaisemartin.bsky.social · May 9

Happy to share videos of 12-hour course on DIDs I taught at @liser.lu in Luxemburg. Covers in depth the first 7 chapters of our textbook, and last chapter more quickly. Use it to teach yourself (for free!) recent developments on DID! Thanks ‪@andreaecon.bsky.social‬ for class and videos!

Andrea Albanese @andreaecon.bsky.social · May 8

🎥 The recording of the Difference-in-Differences course by Prof. @cdechaisemartin.bsky.social , which I had the pleasure of organizing at @liser.lu , is now online!

📚 12 hours covering everything from the basics to advanced methods.

👉 Watch the course here: www.youtube.com/playlist?lis... #EconSky

Difference-in-Differences (DiD) | Training with Prof. Clément de Chaisemartin - YouTube

Join Prof. Clément de Chaisemartin in this in-depth training on Difference-in-Differences (DiD) — a powerful econometric tool for causal inference and policy...

www.youtube.com

2 9 60

Clément de Chaisemartin @cdechaisemartin.bsky.social · Apr 18

Just posted updated version of our DID textbook! We now have drafts of all chapters, including the one on general designs! Now you can tell your friends still on X that they are DID-outdated :-) Happy easter for those of you that celebrate it. papers.ssrn.com/sol3/papers....

Credible Answers to Hard Questions: Differences-in-Differences for Natural Experiments

This book introduces applied researchers to modern Differences-in-Differences (DID) methods, that they can use to obtain credible answers to hard causal inferen

papers.ssrn.com

14 80 260

Reposted by Clément de Chaisemartin

James Bland @jamesbland.bsky.social · Apr 18

🚩New version of a working paper now available!🚩
"Rounding the (Non)Bayesian Curve: Unraveling the Effects of Rounding Errors in Belief Updating" with Yaroslav Rosokha

papers.ssrn.com/sol3/papers....

A thread 👇👇

#Bayes #TeachBE #EconSky

2 3 12

Reposted by Clément de Chaisemartin

LISER - Luxembourg Institute of Socio-Economic Research @liser.lu · Apr 2

Day 2 of the course on Difference-in-Differences (DiD) at the Jane Jacobs Conference Room, LISER, with Prof. Clément de Chaisemartin @cdechaisemartin.bsky.social.

ℹ️ Learn more: www.liser.lu/course-diffe...

Organised by @andreaecon.bsky.social & Pierre Imhoff 🤝

#EconSky #Econometrics #Education

1 8

Reposted by Clément de Chaisemartin

LISER - Luxembourg Institute of Socio-Economic Research @liser.lu · Apr 2

🏁 Day 2 of Prof. Clément de Chaisemartin's visit concludes with a seminar: "Difference-in-Differences Estimators of Intertemporal Treatment Effects"

🙏 Big thanks to @cdechaisemartin.bsky.social and organisers @andreaecon.bsky.social et al.

ℹ️ www.liser.lu/course-diffe...

#EconSky #Econometrics

3 6

Clément de Chaisemartin @cdechaisemartin.bsky.social · Apr 1

Please RT: I am hiring RAs, for 12-month contracts, starting in June/July or early September 2025. Apply if you’re interested in working on some fun and useful econometrics projects!

#EconSky

euraxess.ec.europa.eu/jobs/331499

Research Assistant ERC REALLYCREDIBLE

Clément de Chaisemartin is looking for research assistants, for 12 months contracts, starting in June/July or early September 2025.

euraxess.ec.europa.eu

8 8

Reposted by Clément de Chaisemartin

Andrea Albanese @andreaecon.bsky.social · Mar 31

Last chance to register for our Difference-in-Differences course with Prof. Clément de Chaisemartin (@cdechaisemartin.bsky.social), taking place tomorrow and Wednesday – April 1st & 2nd!

Join us at @liser.lu or online for this free course.

Register now: www.liser.lu/fr/course-di...
#EconSky

testing

LISER's public research focuses on social and economic policies, including their spatial dimension. Its mission is to provide the scientific community and society with clear, relevant and solid answer...

www.liser.lu

7 5

Reposted by Clément de Chaisemartin

LISER - Luxembourg Institute of Socio-Economic Research @liser.lu · Apr 1

👋 A big LISER welcome to Prof. Clément de Chaisemartin (@cdechaisemartin.bsky.social) who has arrived in #Luxembourg to give a 2-day course on Difference-in-Differences (DiD).

ℹ️ Learn more: www.liser.lu/course-diffe...

💪Organised by @andreaecon.bsky.social et al.

#EconSky #Econometrics #Education

1 4 6

Reposted by Clément de Chaisemartin

Andrea Albanese @andreaecon.bsky.social · Mar 28

📘 Ready for Day 2 of our hybrid Difference-in-Differences course at @liser.lu with Prof. Clément De Chaisemartin (@cdechaisemartin.bsky.social)

Still possible to join for Days 3 & 4 (April 1–2, 2025)! Free online registration is open:
👉 www.liser.lu/course-diffe...
#EconSky #CausalInference

3 5

Reposted by Clément de Chaisemartin

Andrea Albanese @andreaecon.bsky.social · Mar 25

📢 Excited to kick off our 12-hour hybrid course on Difference-in-Differences today, led by Prof. Clément De Chaisemartin (@cdechaisemartin.bsky.social) at LISER.

Still time to join!

Free online registration open for Days 3 & 4 (April 1–2, 2025): www.liser.lu/fr/course-di.... #EconSky

7 7

Reposted by Clément de Chaisemartin

LISER - Luxembourg Institute of Socio-Economic Research @liser.lu · Mar 25

⭐LISER is honoured to host a 12-hour course on Difference-in-Differences, led by Prof. Clément de Chaisemartin, starting today.

ℹ️ Free registration for Days 3 and 4 is still possible!

ℹ️ Learn More & Register here:
www.liser.lu/course-diffe...
.
@andreaecon.bsky.social
@cdechaisemartin.bsky.social

Difference-in-Differences

LISER's public research focuses on social and economic policies, including their spatial dimension. Its mission is to provide the scientific community and society with clear, relevant and solid answer...

www.liser.lu

2 5

Clément de Chaisemartin @cdechaisemartin.bsky.social · Mar 12

Dofile and datasets to replicate our empirical applications in Chapters 3 and 4 are available from SSC:

ssc desc cc_xd_didtextbook
net get cc_xd_didtextbook

2 2 12

Clément de Chaisemartin @cdechaisemartin.bsky.social · Mar 12

Chapter 5 on designs with variation in treatment timing is in good shape. Chapter 6 on designs with variation in treatment dose is also in good shape. So what's left is essentially Chapter 7, on general designs.

1 1 7

Clément de Chaisemartin @cdechaisemartin.bsky.social · Mar 12

Slowly but surely coming along: we have a new version of our working textbook on diffs in diffs!

papers.ssrn.com/sol3/papers....

Chapters 2 to 4, which cover the set up, classical DIDs, and relaxations of the parallel trends assumptions have been thoroughly revised and are now almost finished.

Credible Answers to Hard Questions: Differences-in-Differences for Natural Experiments

This book introduces applied researchers to modern Differences-in-Differences (DID) methods, that they can use to obtain credible answers to hard causal inferen

papers.ssrn.com

5 39 150

Clément de Chaisemartin @cdechaisemartin.bsky.social · Mar 3

We have a 4th Stata package, multisit_regLATE, to estimate sign of reg coefficients of sites' LATEs on characteristics. For instance, can be used to test if LATEs positively correlated to sites' first stages (consistent w Roy model). ssc install multisite, then run it once to install 4 packages.

Clément de Chaisemartin @cdechaisemartin.bsky.social · Feb 13

@Antoinedeeb.bsky.social and I have new paper and Stata commands, to estimate & predict treatment-effect heterogeneity in multi-site RCTs! papers.ssrn.com/sol3/papers....
Estimators computed by multisite_varITT multisite_regITT and multisite_varLATE Stata packages, available from SSC.

1 10