This course offers an introduction into causal inference with directed acyclic graphs (DAGs). DAGs combine mathematical graph theory with statistical probability concepts and provide a powerful approach for causal modeling. Originally developed in the computer science and artificial intelligence field, they recently gained increasing traction also in other scientific disciplines (such as economics, political science, sociology, health sciences, and philosophy). DAGs allow to check the validity of causal statements based on intuitive graphical criteria, that do not require algebra. In addition, they open the possibility to completely automatize the causal inference task with the help of special identification algorithms. As an encompassing framework for causal reasoning, DAGs are becoming an essential tool for everyone interested in data science and machine learning. The course provides a good overview of the theoretical advances that have been made in the field of causal data science in the last thirty years. The focus lies on practical applications of the theory and students will be put into the position to apply the covered methodologies in their own research. In particular, common causal inference challenges such as backdoor adjustment, bad controls, instrumental variables, selection bias, and external validity will be discussed in one single framework. Hands-on examples using dedicated libraries in R will guide through the presented material. There are no prerequisites for participating, but a good working knowledge in basic statistics and R are a plus.

 

A detailed syllabus with course times and literature will soon be available for download here.

By the end of the course participants will: