Entry requirements

120 credits including 90 credits in statistics.

Learning outcomes

A student who has taken this course will:

show an in-depth knowledge of the potential outcomes framework and use of directed acyclic graphs for causal inference,

show an in-depth understanding of assumptions underlying causal analyses with experimental and observational data,

master theory presented for estimation of causal parameters in randomized experiments and observational studies,

master the application of parametric and non/semiparametric estimators of causal effects presented in the course,

be able to perform sensitivity analyses on estimates of a causal effect.

Content

Potential outcomes, theory and assumptions

Fisher's exact test and Neymans approach to completely randomized experiments

Causal effect estimators, propensity score - model building, stratification and matching

Variance estimation

Instrumental variable methods

Directed acyclic graphs (causal DAGs), construction, application and interpretation

Mediation analysis - parameters - estimatorsSensitivity analysis and boundsUndervisning

Instruction

Instruction is given in the form of in-class lectures and computer exercises.

Assessment

The examination takes place through a written examination and/or through written and/or oral presentation of take-home assignments.

Other directives

The course is included in the Master´s programme in statistics