The first issue of The Journal of Network Theory in Finance for 2017 contains two papers that look at market microstructure and how its features can be used as early warning signals for market troubles. It also features two papers that investigate the very different types of contagion in financial markets.
The issue’s first two papers were presented at the twelfth edition of the Econophysics Colloquium, which was held last year in São Paulo, Brazil.
In “Quantifying the diversity of news around stock market moves”, Chester Curme, Ying Daisy Zhuo, Helen Susannah Moat and Tobias Preis analyse whether the news being dominated by only a few big stories or by more diverse ones, covering a multitude of topics, has an impact on the performance of financial markets. The authors find that the diversity of news both responds to and influences the behavior of financial markets. In particular, the method they propose can improve forecasts of trading volume and offer early warning signals of increased activity in financial markets.
Research in recent years has shown that many of the stylized facts of financial markets are not as stylized when evaluated quantitatively over longer periods of time. Our second paper, “Nonstationarity of the intraday individual and collective seasonalities of price fluctuations” by Sílvio M. Duarte Queirós and Michelle B. Graczyk, demonstrates that this is also the case for the [-shaped curve of trading volume across a day, where most trading volume is expected to take place in the morning and afternoon.
The authors extend this analysis to look at returns and correlations of returns, and they find that periods relating to the subprime and debt crises have distinct [-shapes. Such features of market microstructure could be used as risk indicators for the market. The third paper in this issue, “How the interbank market becomes systemically dangerous: an agent-based network model of financial distress propagation” by Matteo Serri, Guido Caldarelli and Giulio Cimini, develops an agent-based model for banks’ overnight lending activity. Using real balance sheet data for 183 European banks, the authors find extreme fragility in the market leading up to the 2007–8 financial crisis. Agent-based models like this one could be run on a regular basis to assess the evolution of the riskiness of the system. While a vast body of literature exists on modeling contagion, on both the hard asset and liability sides of the balance sheet (as in our third paper), very little research has measured contagion in relation to a firm’s goodwill item on its balance sheet.
Our fourth paper, “Reputation risk contagion” by Peter Mitic, measures howthe reputation of a bank depends on how its customers see its reputation co-move with that of other banks. The author finds that 10 15% of the reputation of any bank is attributable to such network effects. This raises the question of how banks can actively shape their influence networks to be more advantageous with regard to the spread of a positive reputation from other banks, while neutralizing the spread of a negative reputation.
Finally, we are pleased to announce that The Journal of Network Theory in Finance has been accepted for indexing in the Emerging Sources Citation Index produced by Thomson Reuters. Journals selected for this new index gain the benefits of content discoverability via Web of Science while continuing to be monitored and considered for potential future inclusion in the Social Sciences Citation Index. For more information, please visit http://wokinfo.com/products_tools/multidisciplinary/esci/.
Kimmo Soramäki and Tiziana Di Matteo
Financial Network Analytics Ltd. and King’s College London
In this paper, the authors use a topic-modeling approach to quantify the changing attentions of a major news outlet, the Financial Times, to issues of interest.
This paper deals with statistical measures based on high frequency data from stock markets, and in particular looks at how these measures changed according to time, with a focus on before and after the crisis of 2008.
How the interbank market becomes systemically dangerous: an agent-based network model of financial distress propagation
In this paper, the authors study the stability of the interbank market to exogenous shocks using an agent-based network framework.
The aim of this paper is to assess the effects of the reputation of the members of a group on any single member of the group using the concepts of social influence and convergence in belief.