A Complex Network Approach to Identify Potential Financial Scandals: The Colombian Market Case

Financial data is abundant, diverse and generated in large volumes at any time worldwide. Finding fast and reliable ways of analyzing it is key for market actors (regulators, market makers, brokers and investors). In this work, we intend to use graph theory as a vehicle to analyze relationships among owners of publicly traded companies in order to extract latent dynamics that are very difficult to discover otherwise. As a case study we took the bankruptcy of the largest brokerage firm in Colombia in 2012. Network theory uncovered latent dynamics in the relationships among key owners and companies during the period of analysis (2009 2012) that could be used in the future as an early warning tool for market participants.