Online Program Home
My Program

Abstract Details

Activity Number: 544 - Dynamic Graphical Models and Networks with Applications
Type: Invited
Date/Time: Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
Sponsor: International Indian Statistical Association
Abstract #300467
Title: Network Modeling of High-Dimensional Time Series
Author(s): Sumanta Basu*
Companies: Cornell University
Keywords: time series; quantile regression; spectral density; graphical model; lasso; vector autoregression

Estimating network connectivity among multiple time series is an important problem in many areas of biological and social sciences. Examples include system-wide risk monitoring in financial economics and functional connectivity analysis in neuroscience. In this talk, I will present some recent works on building networks from multivariate time series data, where the core idea is to learn from lead-lag relationships between two time series in addition to their contemporaneous association. In the context of modeling networks of interactions among financial firms and measuring systemic risk, we will demonstrate how linear and quantile-based Granger causality analyses using vector autoregressive (VAR) models can provide insight. We will conclude with some model-free techniques for building networks among time series using multivariate spectral density and its inverse, and show their applications in functional connectivity analysis. Asymptotic theory of these methods under high-dimensional regime will be discussed.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program