Servicenavigation

Archivierte Version! ⇒ Zur Aktuellen [de]…      Archived Version! ⇒ Go to Current [en]…
Statistik: mehr als Erbsen zählen

You are here:

Lecturer

Seminar "Causal Inference in Statistics and Machine Learning"

Description

Although most studies in the life and social sciences aim to answer causal questions, their approach is usually focussed on associations.

Typical causal questions are:

  • What was the death cause of a given individual, in a specific incident?
  • What fraction of crimes could have been avoided by a given policy?
  • What is the effect of smoking on mortality in a given population?

The aim of this seminar is to jointly work out some of basic concepts and ideas of causal inference and to understand methods for quantifying causal effects.

Literature

  • Hernan & Robins (2020). Causal Inference - What if. CRC Press.
  • Peters, Janzing & Schölkopf (2017). Elements of causal inference. The MIT Press.

Requirements

  • This seminar is eligible for Master students in Statistics, Data Science and Econometrics.
  • Some of the early topics are suitable for interested statistics students in their last Bachelor year.
  • A decent understanding of probabilistic and statistical concepts and linear algebra is expected.

Mode

  • The seminar will take place over the course of 2-4 block meetings starting around the middle of the semester.
  • Grading will be based on a written report (either English or German) and a presentation (English).

Registration

Binding registration is possible via e-mail to sia.statistik@tu-dortmund.de with matriculation number and study program.