A0554
Title: Estimating time-varying networks for high-dimensional time series
Authors: Yuning Li - University of York (United Kingdom) [presenting]
Abstract: Time-varying networks are explored for high-dimensional locally stationary time series, using the large VAR model framework with transition and (error) precision matrices evolving smoothly over time. Two types of time-varying graphs are investigated: one containing directed edges of Granger causality linkages and the other containing undirected edges of partial correlation linkages. Under the sparse structural assumption, a penalized local linear method is proposed with time-varying weighted group LASSO to jointly estimate the transition matrices and identify their significant entries and a time-varying CLIME method to estimate the precision matrices. The estimated transition and precision matrices are then used to determine the time-varying network structures. Under some mild conditions, the theoretical properties of the proposed estimates are derived, including the consistency and oracle properties. In addition, the methodology and theory are extended to cover highly correlated large-scale time series, for which the sparsity assumption becomes invalid, and it is allowed for common factors before estimating the factor-adjusted time-varying networks. Extensive simulation studies and an empirical application are provided to a large U.S. macroeconomic dataset to illustrate the finite-sample performance of the methods.
