Research interests:
Causal inference; Machine learning; Econometrics; Digitization and data-driven decision-making in marketing, health, and public policy.
Working papers
Detecting and Mitigating Algorithmic Bias in Heterogeneous Treatment Effects: Theory, Methods, and Empirical Evidence
Joel Persson, Jurrien Bakker, Dennis Bohle, Stefan Feuerriegel and Florian von Wangenheim
[draft] [video]
Conference presentations:Workshop on Platform Analytics 2024, San Diego, April 6-8
INFORMS Marketing Science Conference 2023, Miami, June 8-10, 2023
13th Annual Theory + Practice in Marketing (TPM) Conference (2023), Lausanne, May 29-31. 2023
The 2023 American Causal Inference Conference (ACIC), Austin, May 24-26. 2023
Marketing Science: Diversity, Equity And Inclusion Conference @ SMU Cox, Dallas, March 24–25, 2023
Seminar presentations:
PhD Seminar in Quantitative Marketing Research, University of Zurich, March 7, 2023
CER-ETH Brown Bag Lunch Seminar, Center of Economic Research at ETH Zurich, March 1, 2023
Machine Learning Foundations Department, Booking.com, Nov 26, 2022
Off-Policy Learning of Audience Content Promotions
Joel Persson, Stefan Feuerriegel and Cristina Kadar
[working paper]
(previously presented under other titles)
Conference presentations:
2023 INFORMS Annual Meeting, Phoenix, Oct 15–18, 2023; Session Chair for "Machine Learning and Optimization Applications"
QME conference 2023, Imperial Business School, Sept 1–2 2023
5th Workshop of the EURO Working Group on Pricing and Revenue Management (EURO Workshop 2023), University of Zurich, Aug 28–29, 2023
INFORMS Revenue Management and Pricing Section Conference 2023, Imperial Business School, July 9–12, 2023
INFORMS MSOM2023, McGill University, June 25–26, 2023
Causal Data Science Meeting 2021 Virtual, Nov 15–16, 2021
Statistical Challenges in Electronic Commerce Research (SCECR 2021), Virtual, June 17–18, 2021
ISMS Marketing Science Conference 2021, Virtual, June 3–5, 2021
Poster presentations:
ACM Conference on Economics and Computation (EC'23), King's College, June 19–23, 2023
Seminar presentations:
PhD Seminar in Quantitative Marketing Research, University of Zurich, March 7, 2021
Invited presentations:
AI Keynotes, Institute of Artificial Intelligence (AI) in Management, LMU Munich, Virtual, Feb 11
World Association of News Publishers (WAN-IFRA), Virtual, Nov 21, 2023 [link to presentation]
Publications
Monitoring the COVID-19 Epidemic with Nationwide Telecommunication Data
Joel Persson, Jurriaan Parie and Stefan Feuerriegel, "
Proceedings of the National Academy of Sciences, Jun 2021, 118 (26)
[paper] [working paper version] [online supplements] [GitHub]
Invited talks:
Public Health Agency of Canada, Virtual, Feb 28, 2021
National Institute of Statistics and Economic Studies in Luxemburg (STATEC), Virtual, Jan 26, 2021 [slides]
Poster presentation:
AI+X Summit, ETH AI Center, Zurich, Oct 15, 2021
Media coverage:
Recent media coverage – Chair of Management Information Systems | ETH Zurich
CNBC, The Telegraph, Florida News Times, The Greater India, Radio Canada, Bild, LeFigaro, SPIEGEL Wissenschaft, Times of Malta, Telebasel, Hamburger Abendblatt, Beobachter, hna.de, Frankfurter Neue Presse, St. Galler Tagblatt, watson.ch, Suddeutsche Zeitung, spektrum.de, rnd.de, Tages-Anzeiger, Neue Zurcher Zeitung, tagesschau.de, SWI, ZDF heute, orf.at, faz.net, n-tv.de, Schweiner Volkszeitung, t-online.de, oe24.at, Schwabische Ravensburg, web.de, Neue Osnabruckner Zeitung, Potsdamer Neueste Nachrichten, SRF, Tagesschau, Hauptausgabe, bluewin.ch, Sudkurier, Merkur, Redaktionsnetzwerk, thelocal.ch, lenews.ch, Al Arabiya, Le Temps, LesEchos, Lematin, France3, Le Vif, 20.min, L'Express, RTL 5 minutes, atlantico, La Libre, RTS, LaDepeche, SudOuest, and others.
(Most media coverage of any research ever published at the Department of Management, Technology, and Economics at ETH Zurich)
Estimating the effect of mobility on SARS-CoV-2 transmission during the first and second wave of the COVID-19 epidemic, Switzerland, March to December 2020
Adrian Lison, Joel Persson, Nicolas Banholzer and Stefan Feuerriegel,
Eurosurveillance, 2022; 27(10)
[paper] [working paper version] [online supplements]
Learning Optimal Dynamic Treatment Regimes Using Causal Tree Methods in Medicine
Theresa Blümlein, Joel Persson, and Stefan Feuerriegel
Proceedings for Machine Learning for Healthcare (MLHC 2022), PMLR 182:1-25
[paper] [working paper version] [GitHub]
Conference presentations:
Machine Learning for Healthcare (MLHC 2022) Virtual, Aug 5–6, 2022
Chair of Medical Informatics, University of Zurich, Virtual, Feb 11
4. Predicting COVID-19 Spread from Large-Scale Mobility Data
Amray Schwabe, Joel Persson, and Stefan Feuerriegel,
Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining: 3531–3539, 2022
[working paper version]
Conference presentations: KDD 2021, Virtual, Aug 14–18, 2021
Other publications
7 ways to invest in better digital technology to fight pandemics
Joel Persson and Stefan Feuerriegel
World Economic Forum, Jul 7, 2021
[article]
PhD Thesis
Causal Machine Learning for Data-Driven Decision-Making: Methods and Applications
[link]
In my doctoral dissertation, I develop and apply methods at the intersection of causal inference, statistical learning and offline reinforcement learning to address empirical problems in digitization. In particular, I study dynamic treatment personalization in digital health, optimal content promotions for online news publishers, and the detection and mitigation of algorithmic bias in heterogeneous treatment effect models deployed on online platforms. To this end, I adapt causal machine learning techniques to address challenges for their use in real-world applications.
Master's Theses
Variable Selection for Estimating Optimal Sequential Treatment Decisions Using Bayesian Networks
Joel Persson
Lund University, Department of Statistics, 2020
[thesis]
Counterfactual Prediction Methods for Causal Inference in Observational Studies with Continuous Treatments
Joel Persson
Lund University, Department of Statistics, 2019
Conference presentations: Swedish Statistical Society, Stockholm, May 2019
[thesis]