Events in Physics

Dynamic Communities in Multichannel Data

Network communities are sets of nodes in a network that are connected to each other more than they are to
the rest of the network. We investigate the clustering dynamics of multichannel (multivariate) time series
by first representing their correlations as time-dependent networks and then examining the evolution of
network communities. To do this, we employ a node-centric approach that allows us to track the functional
roles of individual nodes in time without having to track entire communities. As an example, we consider a
foreign exchange market network in which each node represents an exchange rate and each edge represents a
time-dependent correlation between the rates. Using dynamical community detection, we find that exchange
rates with strong intra-community connections are persistently assigned to communities with the same set of
nodes.

Work with Dan Fenn and Mason Porter