Experimental Design and Causality

Enrollment Requirements: Graduate Prerequisite: CDSDX 603 This module focuses on the essential distinction between predictive modeling and causal inference in data science. Learners gain an understanding of situations where predictive models may fall short in revealing underlying causality. Real-world examples underscore the potential pitfalls of relying on simple correlations, emphasizing the necessity of experimentation. Learners will gain an understanding of the foundational principles of causal inference, including potential outcomes and counterfactuals. The module explores the principles and applications of A/B testing for methodically assessing the impact of interventions and changes. Students develop practical skills, designing and implementing basic A/B tests, selecting appropriate metrics, and determining sample sizes.