This course provides a rigorous overview of the statistical foundations of causal inference and introduces modern analytic methods for estimating causal effects in randomized trials and observational studies. Topics include outcome regression, propensity-score methods, doubly robust estimators, instrumental variables, approaches for estimating heterogeneous treatment effects (useful for precision medicine), marginal structural models to handle time-varying confounding, and sensitivity analyses for unmeasured confounding; the course also covers study-design considerations such as estimand choice and adaptive randomization. BMDS 250 on clinical trial design is a helpful complement but not required. Prerequisites: working knowledge of statistical inference, probability theory, and R.In addition, for both BMDS 250 and BMDS 251, instructors should be Ying Lu (PI) and Lu Tian (PI).
3 units · Medical Option (Med-Ltr-CR/NC)
This course provides a rigorous overview of the statistical foundations of causal inference and introduces modern analytic methods for estimating causal effects in randomized trials and observational studies. Topics include outcome regression, propensity-score methods, doubly robust estimators, instrumental variables, approaches for estimating heterogeneous treatment effects (useful for precision medicine), marginal structural models to handle time-varying confounding, and sensitivity analyses for unmeasured confounding; the course also covers study-design considerations such as estimand choice and adaptive randomization. BMDS250 on clinical trial design is a helpful complement but not required. Prerequisites: working knowledge of statistical inference, probability theory, and R.In addition, for both BMDS 250 and BMDS251, instructors should be Ying Lu (PI) and Lu Tian (PI).
Offered in Winter 2026 at Stanford University.