Document Type
Conference Paper
Rights
Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence
Disciplines
1.2 COMPUTER AND INFORMATION SCIENCE, Computer Sciences, Information Science, Bioinformatics
Abstract
In this paper we analyse how the behavior of an online financial community in time of geopolitical crises. In particular, we studied the behaviour, composition and communication patterns of online investors before and after a military geopolitical event. We selected a set of 23 key-events belonging to the 2003 US-led invasion of Iraq, the Arab Spring and the first period of the Ukraine crisis. We restricted our study to a set of eight so called military stocks, which are US-manufacturing companies active in the defence sector. We studied the resilience of the community to information shocks by comparing the community composition, its sentiment and users’ communication networks before and after an event at different time intervals. We found how community reaction is governed by ordered patterns. Experimental evidence suggested how in the aftermath of an event the community does not lose its information sharing functionality. Communication networks show a higher in-degree Gini index, connectivity and a rich-club effect. Discussions tend to develop around central users acting as hubs. These backbone users correspond to rich-club users, present both before and after an event, whose sentiment is less volatile than other users and that were previously recognized as local experts of a specific stock. As further evidence of community resilience, the equilibrium of all the indicators analysed is restored after two weeks.
DOI
https://doi.org/10.1145/3106426.3106510
Recommended Citation
Usher, J. & Dondio, P. (2017). Analysing the Behaviour of Online Investors in times of Geopolitical Distress: A Case Study on War Stocks. Proceeding WI '17 Proceedings of the International Conference on Web Intelligence, pg. 275-283. doi:10.1145/3106426.3106510
Publication Details
Published in: · Proceeding WI '17 Proceedings of the International Conference on Web Intelligence Pages 275-283
Leipzig, Germany — August 23 - 26, 2017