Financial Networks: A Study of the Toronto Stock Exchange

In this study, filtered network approaches such as Minimum Spanning Tree and Planar Maximally Filtered Graphs, are used to analyse the topological structure of constituents of S&P Toronto Stock Exchange Composite Index for a period of three years from January 1, 2015 till Decemeber 31, 2017. For this purpose, rolling correlation for each pair of stocks was calculated for six different time windows of 1, 2, 3, 4, 6 and 12 months. Based on the topological structure, the stocks were categorized into core and peripheral stocks using network measures such as degree centrality, betweenness centrality, eccentricity and eigenvector centrality. Categorization of stocks into core and peripheral was consistent for both MST and PMFG based networks for all time windows. Financial stocks were found to be core stocks. Topological structure helps to understand the inter-relationship among the stocks. It would aid in interpreting the nature of economic factors affecting similar group of stocks. Identification and categorisation of core and peripheral stocks could be used as a base for construction of portfolios and risk management.

[1]  R. Prim.  Shortest connection networks and some generalizations , 1957 .

[2]  Hamid Reza Karimi,et al.  Prediction of stock index futures prices based on fuzzy sets and multivariate fuzzy time series , 2015, Neurocomputing.

[3]  A. H. Shirazi,et al.  Network analysis of a financial market based on genuine correlation and threshold method , 2011 .

[4]  Surendra S. Yadav,et al.  A PCA-DEA framework for stock selection in Indian stock market , 2017 .

[5]  Tiziana di Matteo,et al.  Centrality and Peripherality in Filtered Graphs from Dynamical Financial Correlations , 2008, Adv. Complex Syst..

[6]  Štefan Lyócsa,et al.  Stock market networks: The dynamic conditional correlation approach , 2012 .

[7]  R. Mantegna.  Hierarchical structure in financial markets , 1998, cond-mat/9802256.

[8]  J. Kruskal.  On the shortest spanning subtree of a graph and the traveling salesman problem , 1956 .

[9]  Benjamin Miranda Tabak,et al.  Quantifying price fluctuations in the Brazilian stock market , 2009 .

[10]  Frank J. Fabozzi,et al.  Trade the tweet: Social media text mining and sparse matrix factorization for stock market prediction , 2016 .

[11]  Francis C. M. Lau,et al.  A network perspective of the stock market , 2010 .

[12]  T. Aste,et al.  Spread of risk across financial markets: better to invest in the peripheries , 2013, Scientific Reports.

[13]  P. Samuelson.  Proof that Properly Anticipated Prices Fluctuate Randomly , 2015 .

[14]  Kenneth Steiglitz,et al.  Combinatorial Optimization: Algorithms and Complexity , 1981 .

[15]  Ayse Basar Bener,et al.  Topic Detection and Document Similarity on Financial News , 2018, Canadian Conference on AI.

[16]  F. Pozzi,et al.  Dynamical correlations in financial systems , 2007, SPIE Micro + Nano Materials, Devices, and Applications.

[17]  M Tumminello,et al.  A tool for filtering information in complex systems. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[18]  S. Ross.  The arbitrage theory of capital asset pricing , 1976 .

[19]  Ravi Shankar,et al.  Discrete Wavelet Transform-Based Prediction of Stock Index: A Study on National Stock Exchange Fifty Index , 2015, 1605.07278.