A preliminary schedule is:
- Introduction, potential outcomes, basic counterfactual model
- Horvitz-Thompson, standardisation, and double robust estimator of mean causal effect, asymptotics, semiparametric
- Other methods: standardisation, matching, likelihood, marginal models
- Graphical models (undirected, directed, mixed)
- Causal graphs, nonparametric structural equations, interventions, SWIGS, adjustment sets, special cases(mediation, longitudinal studies, and/or linear models)
- Instrumental variable regression, proximal inference
- Causal discovery (faithfulness, Markov equivalence, testing methods
We provide full lecture notes to the course.