Detection and clustering of lead-lag networks for multivariate time series with an application to financial markets

Conference item : Abstract

In this paper, we propose a method for the detection of lead-lag clusters in multivariate time series, using a pairwise lead-lag metric and a directed network clustering algorithm. We demonstrate that the latent network of pairwise lead-lag relationships between time series can be helpfully construed as a directed network, for which there exists a suitable algorithm for the detection of pairs of lead-lag clusters with high pairwise imbalance. Our method is able to detect statistically significant lead-lag clusters in our primary domain of study, the US equity market. We study the nature of these clustersin the context of the empirical finance literature on lead-lag relations.

Authors: Bennett, S; Cucuringu, M; Reinert, G

• Publication status: Published
• Peer review status: Reviewed (other)
• Publication date: 2022-01-01
• Acceptance date: 2021-07-02
• Event title: 7th Workshop on Mining and Learning from Time Series (MiLeTS) at KDD 2021
• Event location: Virtual event
• Event website: https://kdd-milets.github.io/milets2021/
• Event start date: 2021-08-14
• Event end date: 2021-08-18
• Language: English
• Keywords: FFR
• Subtype: Abstract