Granger Causality Networks: Causal Inference for Time Series Data

Speaker : Ryan J. Kinnear
University of Waterloo
Date: 06/12/2016
Time: 2:00 pm - 3:00 pm
Location: LINCS Meeting Room 40

Abstract

Identifying causal (rather than merely
correlative)relationships in physical systems is a difficult task,
particularly ifit is not feasible to perform controlled experiments.
Granger’snotion of causality was developed first in economics beginning
in the1960s and can be used to form a network of “plausible
causalrelations” given only the opportunity to observe the system.
Thismethod is applied, for example, in neuro-imaging to
identifyrelationships amongst brain regions, and in biostatistics to
exploregene regulatory networks. In this talk, we provide an overview
of thenotion of Granger Causality, some methods for learning
GrangerCausality Networks in practice, and our current directions
forresearch.