A1050
Title: Network estimation using graphical lasso
Authors: Maddie Shelley - University of York (United Kingdom) [presenting]
Chiara Boetti - University of Bath (United Kingdom)
Matthew Nunes - University of Bath (United Kingdom)
Marina Knight - University of York (United Kingdom)
Abstract: Network discovery from temporal data observed over the graph nodes has so far been investigated in both time- and spectral (Fourier)-domains, typically under the limiting assumption of weak-stationarity. The purpose is to take a spectral multiscale approach, useful in many practical contexts, under a framework that departs from stationarity. The proposed approach and its success in recovering the process dependence structure over a variety of simulated and real data scenarios are illustrated.
