Falco Bargagli-Stoffi
Potential outcomes; randomization and experiments; observational study design, estimation, and sensitivity; causal machine learning; causal graphical approach.
Sijia Li
Introduction to data fusion (to the theory and practice of combining data sources together to address questions of interest). Special emphasis to applications in public health, medical sciences, policy learning, and machine learning. Specific techniques covered will include test-then-pool strategies, inverse probability weighting, bias correction methods, semiparametric estimation approaches, and Bayesian dynamic borrowing.
Onyebuchi Arah
Introduction to causal inference frameworks, g-methods, mediation analysis, bias analysis for confounding, selection bias, and measurement error.
Roch Nianogo
Quasi-experimental methods: instrumental variable analysis, regression discontinuity design, interrupted time series, difference-in-differences, propensity score matching, synthetic control methods.
Roch Nianogo
Simulation for health decision making, decision analysis, Markov state-transition models, systems dynamics modeling, microsimulation modeling, agent-based modeling, model evaluation.
Onyebuchi Arah
Causal roadmap and more in-depth coverage of g-methods, doubly robust estimation, mediation analysis.
Onyebuchi Arah
Causal identification strategies: (1) backdoor, including variable selection, (2) front-door criterion and formula, (3) instrumental variable analysis including Mendelian randomization and quasi-experimental methods; triangulation.
Elizabeth Rose Mayeda
Lifecourse epidemiology from a causal perspective: DAGs and causal inference frameworks, mediation, mechanisms in lifecourse epidemiology, confounding, measurement error, selection bias, trials and natural experiments, negative controls, simulations for quantifying bias.
Chad Hazlett
Causal assumptions and methods, focus on observational settings.
Chad Hazlett
Study builds conceptual and formal analytical skills required to evaluate causal claims through a focus on conceptual work and more technical analytic methods. Reading and critiquing studies and reporting of studies at a conceptual level. Use of online tools to review statistics and R-coding. Improvement of conceptual understandings by examining the statistical tools used to understand, work with, and make causal claims in a variety of circumstances.
Chad Hazlett
Introduction to causal inference frameworks (potential outcomes, DAGs and do-calculus), conditioning methods of adjustment (matching, propensity scores, weighting, outcome modeling, doubly robust approaches), additional identification strategies (IV, front-door, quasi-experimental methods), partial identification and sensitivity analysis.
Drago Plečko
Identification, estimation, fairness in medical prediction, and reproducible workflows in R and Python.