Interconnectedness in the global financial market

The global financial system is highly complex, with cross border interconnections and interdependencies. As such, financial shocks and events can easily spillover and propagate throughout the entire system. In this highly interconnected environment, local events can be easily amplified and turned into global events. Thus, new models are needed to capture the structure of the global financial village and uncover channels of spillover and contagion. In this paper we analyze the dependencies between almost 4.000 stocks from 15 different countries. We normalize the returns by the estimated volatility using a GARCH model and then use a robust regression process to estimate pairwise dependencies between stocks from different markets. The estimation results are then used to derive network representations, both on the individual and sectoral level. We show that countries like the US and Germany are in the core of this global stock market. Furthermore, we find that the energy, materials and financial sector play an important role in connecting markets, and that this role has been increasing in time for the two former sectors, versus the latter. Thus, the presented framework provides the means to monitor interconnectedness in the global financial system on different aggregation levels, and how they evolve in time.