ORIE 5752

Foundations of Causal Inference for Data-Driven Decisions

Course Description

Course information provided by the 2026-2027 Catalog.

This course provides a rigorous introduction to the theory and practice of causal inference for data-driven decision-making. Students will learn to identify and estimate causal effects using both experimental and observational data, combining statistical foundations with computational tools. Emphasis is placed on understanding causal assumptions, designing valid estimators, and analyzing the theoretical properties of each method. Attention will be given to unique challenges that arise in real-world operational and policy decisions. Example applications include online marketplaces, social networks, and inventory control. Topics include the potential outcomes framework, causal diagrams, randomized and quasi-experimental designs, matching and weighting methods, instrumental variables, difference-in-differences, and introductory causal machine learning.

Prerequisites background equivalent to: ORIE 3500, MATH 2940, and ENGRD 2110 or CS 2110.

Last 4 Terms Offered (None)

Learning Outcomes

• Explain the conceptual differences between correlation, prediction, and causation.
• Model causal relationships using the potential outcomes framework and causal graphs.
• Identify and estimate causal effects from experimental and observational data.
• Apply core causal estimation techniques including regression adjustment, matching, inverse probability weighting, instrumental variables, and difference-in-differences.
• Analyze robustness, sensitivity, and theoretical properties of the above-mentioned casual estimation techniques.
• Interpret and communicate causal findings in decision-making contexts across operations, policy, and analytics.

View Enrollment Information