Statistical Modelling of Downside Risk Spillovers

Abstract

We extend the extreme downside hedge methodology to model sensitivity interconnectedness of market returns to the tail risk of other markets under turbulent conditions. We derive the interconnectedness via Bayesian graph structural learning. The empirical application examines the dynamic interconnectedness among 15 major markets, including G10 economies, during turbulent times. We investigate whether downside risk connections among these major markets are merely anecdotal or provide evidence of contagion and the most central market for spillover propagation. The result shows that the Covid-19 induced downside risk connections record the highest density, suggesting stronger evidence of contagion in the coronavirus pandemic than during the financial and eurozone crisis. Central to the spillover propagation is the finding that most of the transmitters and recipients of downside risk are EU markets.