🎉My book Impact Evaluation in Firms and Organizations turns one! mitpress.mit.edu/978026255292... It introduces impact evaluation/causal inference for business applications in a non-technical way. R/Python code and datasets freely available: www.unifr.ch/appecon/en/r... 📚Lecture slides coming soon!

Impact Evaluation in Firms and Organizations

In today's dynamic business climate, organizations face the constant challenge of making informed decisions about their interventions, from marketing campaig...

mitpress.mit.edu

📖 Nearly 3 years after the publication of Causal Analysis (MIT Press), I'd like to highlight the lecture slides accompanying the book. Available here as PDF and editable TeX files, together with datasets & code in R, Python, & Stata: dataverse.harvard.edu/dataset.xhtm... @mitpress.bsky.social

Had the pleasure of teaching a PhD course on Causal Inference at my alma mater, the Universität Innsbruck, this week. Thank you to all the participants for the engaging discussions-in the stunning heart of the Alps! 🏔️

🚀 Registration is open for the #Fribourg #WinterSchool in #DataAnalytics & #MachineLearning, Feb 1–12 2027. 📍at Fribourg University or online 🔍Topics: data analytics, predictive/causal machine learning, deep learning, agentic AI 💻 Software: Python, R, Knime 👉 Sign up: www.unifr.ch/appecon/en/w...

🎉 Our paper, joint with Yu-Chin Hsu and Yu-Min Yen, is now published open access in JBES. We develop a double machine learning approach to estimate direct and indirect quantile treatment effects, with an application to the National Job Corps Study: doi.org/10.1080/0735... #CausalInference #DoubleML

Estimation of Direct and Indirect Quantile Treatment Effects with Double Machine Learning

We propose a framework to disentangle the quantile treatment effect of a binary treatment at a specific rank into an indirect quantile treatment effect that operates through a mediator and an unmed...

doi.org

📖 News about my book #CausalAnalysis! Many empirical examples are now also available in #Stata - in addition to #R and #Python. You can find the data and code files at 🔗 www.unifr.ch/appecon/en/r... A big thank you to Sarina Oberhänsli for translating the code from R to Stata! 🙏

Very happy to share our new working paper (with D. Imhof & T. Madiès) on cartel behavior, detection, and damages in public procurement: arxiv.org/abs/2606.30470 We show how firms coordinated bids in a Swiss bid-rigging cartel, while mimicking competition. Estimated average overcharges: at least 45%.

Swimming in Dark Water: When Cartels Mimic Competition

This paper analyzes the internal organization and economic effects of a bid-rigging cartel in the road construction sector of the Swiss canton of Ticino, active from 1999 to 2005. Using exceptionally ...

arxiv.org

🔥New working paper out (with Hannes Wallimann & Cédric Brütsch): arxiv.org/abs/2606.24181 Using 21k+ bus inspections in Switzerland, we estimate the causal effect of inspector visibility. Result: uniformed inspections detect ~25% fewer fare evaders per hour than undercover inspections.

Visible or Covert? The Causal Effect of Inspector Visibility on Fare Evasion Detection: A Causal Machine Learning and Policy Learning Approach

Fare evasion generates substantial revenue losses for public transport operators and is typically combated through fare inspections, yet little is known about how the mode of inspection-uniformed vers...

arxiv.org

Delighted to be attending the Econometric Society African Meeting in Cairo (#AFES2026) as invited speaker. Will present joint work with M. Haddad, J. Medina-Reyes, L. Zhang on difference-in-differences for continuous treatments using machine learning: virtual.oxfordabstracts.com/event/76129/...

✅ today: econometrics, treatment effects. 15 days to prelim #econsky #econtwitter book by @causalhuber.bsky.social

🎉Our paper "Testing Monotonicity of Mean Potential Outcomes in a Continuous Treatment with High-Dimensional Data", joint work with Yu-Chin Hsu, Ying-Ying Lee, and Chu-An Liu, has been published in the Review of Economics and Statistics @restatjournal.bsky.social: direct.mit.edu/rest/article...

Testing Monotonicity of Mean Potential Outcomes in a Continuous Treatment with High-Dimensional Data

Abstract. We propose a Cramér–von Mises–type test for testing whether the mean potential outcome given a specific treatment level has a weakly monotonic relationship with the continuous treatment unde...

direct.mit.edu

🆕 New working paper: When is exploiting treatment changes rather than treatment levels valid for causal inference? I develop a formal framework and characterize when strategies based on treatment changes and treatment levels (e.g., DiD) are - and are not - equivalent. arxiv.org/abs/2606.02234

When Do Treatment Changes Identify Causal Effects?

This paper clarifies the identifying assumptions underlying causal inference based on treatment changes rather than treatment levels, and their relationship to conventional identification strategies. ...

arxiv.org

🚀 Excited to share that our paper “A Joint Test of Unconfoundedness and Common Trends”, joint work with Eva-Maria Oeß has now been published in the Journal of Applied Econometrics: onlinelibrary.wiley.com/doi/10.1002/...

A Joint Test of Unconfoundedness and Common Trends

We introduce an overidentification test of two alternative assumptions to identify the average treatment effect on the treated in a two-period panel data setting: unconfoundedness and common trends. ...

onlinelibrary.wiley.com

🔥 Very happy to share our new working paper (with M. Bia, G. Menta, and C. D’Ambrosio) on estimating local average treatment effects using genetic instruments, while allowing for some instruments to be invalid and for heterogeneity in treatment responses across instruments: docs.iza.org/dp18595.pdf

I had the pleasure of visiting IMT in Lucca (www.imtlucca.it), where I presented our recent work on combining difference-in-differences for continuous treatments with double machine learning for control variable selection (joint with M Haddad, J Medina-Reyes, and L Zhang): arxiv.org/abs/2410.21105.

😀 Happy to share our new working paper with Andreas Stoller on cigarette prices, taxes, and smoking: doi.org/10.48550/arX... Using a flexible diff-in-diff approach with double machine learning and Eurobarometer data, we find that tax increases reduce smoking, particularly among the young.

🚀 New working paper with Kevin Kloiber & @lukaslaffers.bsky.social: arxiv.org/abs/2603.04109 We propose a statistical test for full mediation of treatment effects and the identifiability of causal mechanisms, using double machine learning to control for high-dimensional covariates.

Testing Full Mediation of Treatment Effects and the Identifiability of Causal Mechanisms

In causal analysis, understanding the causal mechanisms through which an intervention or treatment affects an outcome is often of central interest. We propose a test to evaluate (i) whether the causal...

arxiv.org

🚀 New working paper - joint with S. Oberhänsli: arxiv.org/abs/2602.23877 We propose a DiD approach to mediation analysis that evaluates direct, indirect, & dynamic treatment effects under conditional parallel trends, using double machine learning for flexible, data-driven covariate control.

Difference-in-differences for mediation analysis using double machine learning

We propose a difference-in-differences (DiD) framework with mediation for possibly multivalued discrete or continuous treatments and mediators, aimed at identifying the direct effect of the treatment ...

arxiv.org

🔥New working paper (joint with A. Armendáriz): We propose a test for the homogeneity of conditional average treatment effects across experimental and observational studies, helping to disentangle unobserved confounding from effect heterogeneity in causal estimates: arxiv.org/abs/2602.19703

Testing Effect Homogeneity and Confounding in High-Dimensional Experimental and Observational Studies

We propose a framework for testing the homogeneity of conditional average treatment effects (CATEs) across multiple experimental and observational studies. Our approach leverages multiple randomized t...

arxiv.org

📣The 2026 Symposium of #CausalInference in the #HealthSciences takes place on March 18, 2026 in Fribourg. Theme: AI & machine learning in causal inference for health sciences 🎤Keynotes: Elsa Gautrain, Aurélien Sallin, Jonas Peters, Jana Mareckova 🔗https://projects.unifr.ch/pophealthlab/?page_id=1561

📢 Registration is open for the 2026 Symposium of Causal Inference in the Health Sciences, hosted at Fribourg University on March 18. This year’s focus is on AI/machine learning in causal inference for health sciences/economics - register here: projects.unifr.ch/pophealthlab... #CausalAI

📢 Last call! Register for the #Fribourg #WinterSchool in #DataAnalytics & #MachineLearning (📅 Feb 2–13, 2026) until Jan 25. Join us on site in Fribourg or online for data analytics, predictive/causal machine learning, and deep learning, using Python, R, Julia, & KNIME: www.unifr.ch/appecon/en/w...

📢 Update of our #DiD paper on continuous treatments with machine learning-based covariate adjustment, joint with M Haddad, J Medina-Reyes, and L Zhang. Now includes an evaluation of the impact of second-dose COVID-19 vaccination rates on mortality in Brazil: arxiv.org/abs/2410.21105

Difference-in-Differences with Time-varying Continuous Treatments using Double/Debiased Machine Learning

We propose a difference-in-differences (DiD) framework designed for time-varying continuous treatments across multiple periods. Specifically, we estimate the average treatment effect on the treated (A...

arxiv.org

The new year starts with a great conference: the Labor Seminar in #Laax 🏔️ Inspiring talks with applications of causal inference methods in empirical labor economics and related fields. Thanks to Pia Schilling and Christina Felfe for putting together a fantastic program!

📄 New paper (joint with J Kueck & M Mattes): arxiv.org/abs/2601.05728 When outcomes depend on others’ actions in a social network, causal evaluation becomes difficult. We use causal AI to learn network interference from data and to test whether common ways of modelling interference are valid.

Learning and Testing Exposure Mappings of Interference using Graph Convolutional Autoencoder

Interference or spillover effects arise when an individual's outcome (e.g., health) is influenced not only by their own treatment (e.g., vaccination) but also by the treatment of others, creating chal...

arxiv.org

📢The #Fribourg #WinterSchool in #DataAnalytics & #MachineLearning is coming up Feb 2-13. Join us on site at @ses_unifr or online for a two-week program on data analytics, predictive/causal machine learning & deep learning using Python, R, Julia & KNIME: www.unifr.ch/appecon/en/w...

Happy holidays from the #Venet in #Tirol, #Austria ❄️⛷️ - the perfect crowd-free ski retreat, recharging energy for fresh #CausalAnalysis and #ImpactEvaluation in the new year 😉 www.venet.at

⏳ Our #Fribourg #WinterSchool in Data Analytics & Machine Learning is only a few weeks away (Feb 2–13, 2026). Strengthen your skills in predictive and causal machine learning, deep learning using Python, R, Julia & Knime. Register here to join us in person or online: www.unifr.ch/appecon/en/w...

Very honored to be recognized as a Distinguished Author of the Journal of Applied Econometrics in 2025 (for the equivalent of three single-authored publications). I’m very grateful to my coauthors - most of my work in this journal has been collaborative! 😊 onlinelibrary.wiley.com/page/journal...

Journal of Applied Econometrics DISTINGUISHED AUTHORS ANNOUNCEMENT

The Journal of Applied Econometrics is a statistical and mathematical economics journal for the application of econometric techniques to economic problems.

onlinelibrary.wiley.com

🚀 A new version of our causalweight package for the statistical software R is online, containing some of the latest causal machine learning methods for the estimation of treatment effects: www.rdocumentation.org/packages/cau... #CausalInference #CausalAnalysis #MachineLearning