This course introduces students to computational and empirical methods for public policy analysis in an AI-assisted world. More than ever, policy professionals must understand technical methods, apply them with sound judgment, and critically evaluate the credibility and limitations of analytical claims. Students will learn to characterize policy problems with data, build and assess predictive models, formulate allocation and optimization problems, evaluate systems, and reason about causal evidence from experiments and observational data. Throughout the course, students will use coding agents to support their work, including in required weekly discussion sections. The emphasis, however, is not on automating analysis, but on developing the judgment needed to ask good questions, check assumptions, diagnose errors, interpret results, and communicate uncertainty and limitations. The goal is to prepare students to use, evaluate, and govern computational systems responsibly in service of public decision-making.
Prerequisites: An introductory course in statistics (e.g., API-201) and some programming experience.