Biostatistics 683 - Intro to Causal Inference

With the recent and ongoing 'data explosion', methods to delineate causation from correlation are perhaps more pressing now than ever. This course will introduce a general framework for causal inference: 1) clear statement of the scientific question, 2) definition of the causal model and parameter of interest, 3) assessment of identifiability, 4) choice and implementation of estimators including parametric and semi-parametric methods, and 5) interpretation of findings. The methods include G-computation, inverse-weighting, and targeted maximum likelihood estimation (TMLE) with Super Learning, an ensemble machine learning method. Students gain practical experience implementing these estimators and interpreting results through discussion assignments, R labs, R assignments, and a final project. Recommended: a graduate course in Epidemiology (e.g. EPI 630).

Prerequisites: NONE. Recommended: (1) graduate-level coursework covering basic probability theory and regression modelling; (2) a graduate course in Epidemiology (e.g. Epi630)

This course is offered in person on the Amherst campus. A synchronous Zoom option may be available for all or selected class meetings; please check with the instructor for details.