Causal Inference in Time Series

Vorlesung im Sommersemester 2024

Prof. Etesami

 

Content

Causal inference is a branch of machine learning and statistics that aims to develop theoretical models and practical algorithms to infer the statistical causal dynamics in complex systems. In this course, we will introduce concepts, principles, and algorithms necessary to solve modern, large-scale problems in scientific inferences, business, and engineering with emphasis on causal inference. Theory and application will be balanced, with students working directly with network data throughout the course.

Tentative list of topics:


Introduction: What is causal inference?
Review of useful preliminaries
Random variable, Time series, Spectral analysis
Sequential predictors,
Causality in Times Series
Granger causality
Directed information graphs
Efficient algorithms
Causal structure learning in time series
Structures with Latent variables
Time-Varying networks
change-point detection
Concrete Applications
Computational neuroscience
Financial markets
Social networks

Slides and exercise sheets

Lecture slides, exercise sheets, and solution hints will only be available in Moodle.