DATA 8020: Advanced Causal Inference

Schedule: Fridays, 2:00 PM-4:50 PM, September 4-November 27, 2026
Venue: HKU IDS P603 Seminar Room, Graduate House
Instructor: Prof. Jin-Hong Du
Offering unit: Institute of Data Science, The University of Hong Kong

Overview

This course introduces modern causal inference from statistical, machine learning, and computational perspectives. Students will learn how to formulate causal questions, identify causal effects under explicit assumptions, and develop statistically principled and computationally scalable methods for estimation and inference from observational and experimental data.

The course combines classical statistical foundations with modern machine learning tools, including double machine learning, high-dimensional modeling, and heterogeneous treatment effect estimation, to prepare students for research in causal methodology and data-rich scientific applications.

Topics

1. Causality in AI and data science
2. Foundations of causal inference
3. Classical design, estimation, and inference
4. Modern machine learning for causal inference
5. Hidden confounding and robustness
6. Computational methods and research frontiers

Learning Outcomes

Upon completion of this course, students will be able to:

1. Formulate research questions in AI and data science as well-defined causal problems.
2. Distinguish between causal identification, estimation, and inference.
3. Apply modern causal inference methods to real-world problems in biomedicine, genomics, the social sciences, and related areas.
4. Implement causal estimators in high-dimensional and computationally challenging settings using reproducible code.
5. Communicate causal assumptions, methods, results, and limitations effectively.

Prerequisites

Basic knowledge of probability, linear regression, and machine learning is expected. Programming experience in Python or R is required.

Assessment

• Final group project — 60% (proposal 10%, presentation 15%, final report 35%)
• In-class quizzes — 30% (about five to six short, closed-book quizzes at the start of lecture; lowest dropped)
• Participation — 10%

Quizzes replace traditional homework: they are individual and closed-book, and verify each student's own mastery of the core material.

Final Project

The final project is the centerpiece of the course. Working in small groups of 1–3, students formulate a causal-inference question and carry it from a written proposal to an in-class presentation and a final research report. Ambitious projects may grow into a workshop or conference submission.

• Proposal — Week 5 (October 2): a brief written proposal, with an optional short in-class presentation.
• Presentations — Weeks 12–13 (November 20 & 27): in-class final presentations.
• Final report — one per group, due in the assessment period (~December 11, 2026; tentative).

Detailed rubrics and a list of candidate projects are in the project handout.

References

Primary textbook:

• Imbens, G. W., and Rubin, D. B. (2015). Causal Inference in Statistics, Social, and Biomedical Sciences. Cambridge University Press.

Supplementary references:

• Hernan, M. A., and Robins, J. M. (2020). Causal Inference: What If. Chapman & Hall/CRC.
• Peters, J., Janzing, D., and Scholkopf, B. (2017). Elements of Causal Inference: Foundations and Learning Algorithms. MIT Press.
• Ding, P. (2023). A First Course in Causal Inference. Chapman and Hall/CRC.
• Chernozhukov, V., Hansen, C., Kallus, N., Spindler, M., and Syrgkanis, V. (2024). Applied Causal Inference Powered by ML and AI. arXiv:2403.02467.

Additional public resources:

• Professor Linbo Wang's teaching page, including Directed Reading: Causal Inference.

Course Materials

HKU Semester 1 teaching begins on September 1, 2026 and ends on November 30, 2026. Lecture notes and homework PDFs will be posted here as they become available.

Syllabus: PDF
Project: Handout

Final report due ~December 11, 2026 (tentative).