Publication:
Resilience of Stock Cross Correlation Network to Random Breakdown and Intentional Attack
Resilience of Stock Cross Correlation Network to Random Breakdown and Intentional Attack
No Thumbnail Available
Files
Date
2017
Authors
Ngoc Kim Khanh Nguyen
Quang Nguyen
Journal Title
Journal ISSN
Volume Title
Publisher
Research Projects
Organizational Units
Journal Issue
Abstract
We study the network constructed by the correlation coefficients between stocks which are listed in the Vietnamese stock exchanges. Network edges between nodes (stocks) are established if the correlations between stocks are higher than certain value (0.25). We found that this network is scale-free, having connectivity distribution P(k)∼k−γ (where k is the node connectivity) with a relatively low power exponent of γ∼1.0. This result accords with the highly co-movement of listed stocks in a market found previously. The low power-law distribution exponent coefficient corresponding to a dense connectivity makes it robust even under intentional attacks: its critical fraction is of range 30.77%−50.36%. Finally, we compare different intentional attack strategies and find that: if we want to fully break apart the network, recalculated degree distribution-based attack is the most efficient; if we only want to break the network to a certain level (for example, to break half of the size of the largest network component), the recalculated betweenness centrality-based attack is more efficient. This result may be used to enhance the structure design of some real-life network systems.