Relation between regional uncertainty spillovers in the global banking system

We report on time-varying network connectedness within three banking systems: North America (NA), the European Union (EU) and Southeast Asia (ASEAN). Diebold and Yilmaz’s original method is improved by using exponentially weighted daily returns as well as ridge regularization on vector autoregression (VAR) and fore- cast error variance decomposition (FEVD). We compute the total network connectedness for each of the three banking systems, which quantifies regional uncertainty. Results over rolling windows of 300 days during the period from January 2005 to October 2015 reveal changing uncertainty patterns that are similar across regions, demonstrating common peaks associated with identifiable exogenous events. Lead– lag relationships among changes of total network connectedness of the three systems, quantified by transfer entropy, reveal that uncertainties in the three regional systems are significantly causally related, with the NA system having the largest influence on the EU and ASEAN.