Experiments and Causal Inference

Course Description

This course introduces experimental and quasi experimental methods for causal inference that are widely used in a broad array of domains such as marketing and information systems. The focus is on delivering a breadth of substantive topics and methodological considerations that emerge in utilizing identification-oriented methods.

Students engage with relevant research papers on each topic, facilitating active discussions about the nature of causation and alternative means of inferring causal relationships.

Final Project

In lieu of a final exam, this course requires students to write a short paper applying or extending the causal inference methods learned in class. It should be no longer than 20 double-spaced pages.

Milestones

• Week 2: Find a collaborator or obtain permission for individual project.
• Week 5: Submit a short (half-page) description of proposed project and plan.
• Week 10: Submit a brief (5 page) memo of main results, including tables/figures.
• Week 16: Submit final version of the project.

Learning Objectives

• Determine which methods and results best support specific empirical inference questions.
• Gain familiarity with causal inference toolkit widely used for business analytics.
• Understand the trade-offs in the design, analysis, and reporting of field, quasi, and natural experiments.

Prerequisites

• Math: Undergraduate-level probability theory and statistics.
• Programming: Proficiency in statistical programming (e.g., R).
• Analysis: Prior experience with regression analysis or econometrics is strongly recommended.