Building directed networks to examine time-series data
Many domains of science deal with time-series data, where the dynamics of many elements are observed over time. Given this data, we can construct models of how these elements effect each-other in time by building directed networks. These models show how activity in one node influences the activity of another node in the future.
In this workshop, Thomas Varley (PhD student,Complex Networks & Systems and Psychology & Brain Sciences, IUB) will cover how information theoretic analyses (transfer entropy, mutual information, etc) can be leveraged to infer the structure of the generative network in a flexible and model-free fashion. We will also discuss the limitations of bivariate network analysis when data can show higher-order (synergistic) relationships.
Prerequisites: a basic understanding of probability theory.
This workshop will be held on-line, with a limited number of seats available in-person. Both attendance options require registration.
Return to the workshop series home page.