A1267
Title: Lead-lag associations in time series: Frequency-specific investigations
Authors: Hernando Ombao - KAUST (Saudi Arabia) [presenting]
Paolo Victor Redondo - King Abdullah University of Science and Technology (Saudi Arabia)
Sipan Aslan - King Abdullah University of Science and Technology (Saudi Arabia)
Abstract: One of the most important facets of time series analysis is the lead-lag dependence structure between components of a system. In brain signals analysis, it is of prime interest to identify how an oscillatory activity (at a particular frequency band) in one channel is associated with an excitation or inhibition at another channel. However, such a lead-lag dependence structure may exhibit some variation across frequency bands. That is, one channel could lead at low frequency but could lag with respect to high frequency. We will explore such behavior through empirical investigations and propose a statistical framework under which we can conduct formal inference on the frequency-specific lead-lag associations between channels (or components) in multivariate time series.
