Analysis of Financial networks in Tehran Stock Exchange using the application of centrality measures
Subject Areas : Financial engineering
Majid Montasheri
1
(PhD Student, Department of Accounting and Finance, Faculty of Management2, Yazd University, Yazd, Iran)
Hojjatollah Sadeqi
2
(Assistant Professor, Department of Accounting and Finance, Faculty of Management2, Yazd University, Yazd, Iran)
Keywords: Clustering, Minimum Spanning Tree, Financial network, Centrality measures,
Abstract :
The purpose of this study is to create a Financial network to identify stock market leaders using centrality measures.This study finally provides a clustering of superior stocks that can be used as an optimal stock portfolio by investors.The statistical population of all stock exchanges is that the 100 companies with the most capital were selected as a statistical sample over a period of 11 years.Due to the nature of the research ranking, the Kendall correlation coefficient was used to calculate the correlation.The Prime algorithm was used to identify relationships and construct the minimum spanning tree, and the fast-greedy algorithm was used to cluster stocks.The results show that in terms of degree centrality measure, stocks of Sepahan Cement companies,Omid Investment Management and Bank Melli Investment, based on closeness centrality measure, stocks of Sepahan Cement companies,International Building Development, and Khuzestan Steel, based on Betweenness centrality measure, the stocks of Sepahan Cement, Ghadir Investment and Bank Melli Investments, and finally based on the bottleneck centrality measure,the shares of Sepahan Cement, Khuzestan Steel and International Building Development have the greatest impact on the stock exchange network. Also, the top stocks were classified into 11 clusters,each of which shows a strong relationship between its components.
_||_
