3.0

Average Course Rating

"Big data" is not necessarily "high quality data." Systematically missing records, unobserved confounders, and selection effects present in many datasets make it harder than ever to answer scientifically meaningful questions. This course will teach mathematical tools to help you reason about causes, effects, and bias sources in data with confidence. We will use graphical causal models, and potential outcomes to formalize what causal effects mean, describe how to express these effects as functions of observed data, and use regression model techniques to estimate them. We will consider techniques for handling missing values, structure learning algorithms for inferring causal directionality from data, and connections between causal inference and reinforcement learning. Pre-requisites: familiarity with the R programming language, multivariate calculus, basics of linear algebra and probability.