Stock market as temporal network

TL;DR

This paper uses temporal network analysis to characterize stock market dynamics, showing that topology evolution can detect instability and that temporal centrality aids in constructing more effective portfolios.

Contribution

It introduces the application of temporal network framework and temporal centrality measures to stock markets for stability detection and improved portfolio optimization.

Findings

Topology evolution indicates market instability.

Peripheral stocks with low temporal centrality outperform in portfolios.

Temporal attributes are crucial for real-world financial applications.

Abstract

Financial networks have become extremely useful in characterizing the structure of complex financial systems. Meanwhile, the time evolution property of the stock markets can be described by temporal networks. We utilize the temporal network framework to characterize the time-evolving correlation-based networks of stock markets. The market instability can be detected by the evolution of the topology structure of the financial networks. We employ the temporal centrality as a portfolio selection tool. Those portfolios, which are composed of peripheral stocks with low temporal centrality scores, have consistently better performance under different portfolio optimization schemes, suggesting that the temporal centrality measure can be used as new portfolio optimization and risk management tools. Our results reveal the importance of the temporal attributes of the stock markets, which should be taken serious consideration in real life applications.