Complex network analysis of the NASDAQ stock market during the initial phase of COVID-19 / Chin Jia Hou and Nor Aziyatul Izni
The United States (U.S.) plays an important role in the global economy, and the COVID-19 pandemic significantly affected the U.S. stock market. Over the past two decades, numerous studies have incorporated complex network analysis to analyze the stock market. However, there is a lack of study focuse...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
Universiti Teknologi MARA, Perak
2024
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/106570/1/106570.pdf https://ir.uitm.edu.my/id/eprint/106570/ http://www.mijuitm.com.my |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The United States (U.S.) plays an important role in the global economy, and the COVID-19 pandemic significantly affected the U.S. stock market. Over the past two decades, numerous studies have incorporated complex network analysis to analyze the stock market. However, there is a lack of study focused on identifying anomalies in the complex network structure of the U.S. stock market that could indicate impending financial crises. The main objective of this research is to implement complex network analysis in examining the changes in the network structures and centralities of the NASDAQ stock networks leading up to and during the initial phase of the COVID-19 pandemic. The opening prices of the stocks under the NASDAQ index in the last two quarters of 2019 and the first quarter of 2020 were collected from Yahoo Finance. The collected data was parsed into edges lists which were then used to construct multiple stock networks. The structures of the stock networks were analyzed using topological metrics such as network density, average clustering coefficient, average path length, network centralizations, and modularity of community structure. The centrality scores of the stocks in the networks were calculated and they were ranked according to the scores. The results show abnormal values in the number of edges, network density, betweenness centralization, and modularity of the community structure during the initial phase of the COVID-19 pandemic. However, no significant anomalies are observed in the average clustering coefficient, average path length, degree centralization, and closeness centralization. Meanwhile, degree centrality proves effective in identifying influential stocks, while closeness and betweenness centralities are found to be less suitable for this particular purpose in the networks used in this study. This paper provides insights into the changes within the stock market at both micro and macro levels around the financial crisis, where the anomalies serve as indicators of an impending financial crisis. |
|---|