Hello! I am an Assistant Professor of Information Systems at the HKUST Business School.

I work on machine learning and causal inference for decision making. Much of my research is about deciding who should receive which intervention—for example, which customers should receive offers, recommendations, or retention efforts. My work shows that better decisions often come not from building better models for a fixed prediction problem, but from choosing a different prediction problem. I study these ideas both theoretically and in collaboration with industry partners.

I earned my PhD in Information Systems from NYU Stern, my MBA from INCAE Business School, and my bachelor's degree in Computer Science from the Tecnológico de Costa Rica.

• Honesty in Causal Forests: When It Helps and When It Hurts (with Yanfang Hou). Accepted at Information Systems Research.
• Observational vs Experimental Data When Making Automated Decisions Using Machine Learning (with Foster Provost). INFORMS Journal on Data Science (2025).
• A Comparison of Methods for Treatment Assignment with an Application to Playlist Generation (with Foster Provost, Jesse Anderton, Benjamin Carterette, and Praveen Chandar). Information Systems Research (2023).
• Evolution of Referrals Over Customers' Life Cycle: Evidence from a Ride-Sharing Platform (with Maxime Cohen and Anindya Ghose). Information Systems Research (2023) .
• Causal Classification: Treatment Effect Estimation vs. Outcome Prediction (with Foster Provost). Journal of Machine Learning Research (2022).
• Causal Decision Making and Causal Effect Estimation Are Not the Same... and Why It Matters (with Foster Provost). INFORMS Journal on Data Science (2022).
• Explaining Data-Driven Decisions made by AI Systems: The Counterfactual Approach (with Foster Provost and Xintian Han). MIS Quarterly (2022).

• Policy Learning for Payment Compliance: Out-of Time Evidence on the Value-and Limits-of ML Targeting at a Brazilian Water Utility (with Felipe A. Araujo, Juliana Dutra, David Hagmann, and Nina Mazar). Under review.
• Causal Ordering Without Effect Estimation: A Framework for Using Proxies in Treatment Prioritization (with Jorge Loría). Working paper.
• Causal Post-Processing of Predictive Models (with Yanfang Hou, Foster Provost, and Jennifer Hill). Working paper.
• Causal Inference Isn't Special: Why It's Just Another Prediction Problem. Opinion piece.

I teach business students how to use machine learning to make better decisions. Students prepare a case before each class and defend their reasoning in discussion; later in the term they build, test, and evaluate their own models in Python. The course also treats generative AI both as a tool students are expected to use and as a subject we examine alongside traditional predictive modeling.

The main course I teach is built around a series of original cases that follow Natalie Ferman, a data-science consultant, and Luis Ortega, her junior analyst, through engagements with clients that include a mobile carrier, a bike-share operator, and a blood bank. The cases turn on questions of judgment—what should be predicted, whether accuracy is the right measure, whether a more complex model is worth it. Students learn the principles behind how models are built, but the course does not work through algorithms one at a time. Four cases from the series are available, along with the current syllabus, which I revise each year.

I am also coauthoring the second edition of Data Science for Business with Foster Provost and Tom Fawcett. We are expanding the new edition to cover generative AI, causal machine learning, and AI strategy and management.

I have been the primary instructor for the following courses:

• 2021–present, HKUST Business School
  • Big Data Analytics^* — MSc in Business Analytics, MSc in Information Systems Management, MSc in Marketing, MBA
  • Data Mining for Business Analytics — undergraduate (2021)
  • AI and Data-Driven Decision Making — executive education
• 2020, NYU Stern — Data Mining for Business Analytics, undergraduate
• 2019, INCAE Business School — Predictive Analytics, Executive MSBA^**

*Despite the name, the course is really about predictive analytics.
**Selected by students as the best instructor in the program.