A multivariate compound dynamic contagion process for modeling spillover dynamics in business and economic sectors

A multivariate process is required to take into account the interconnectedness of business and economic activities. Under a multivariate compound dynamic contagion process, we study the correlation coefficients, moments and conditional value-at-risks (CoVaRs). The derivation of the moments is based on the martingale methodology used by Dassios and Jang (2003). Numerical illustrations are provided for modeling of aggregate losses arising from contagious catastrophic events, such as cyber risk events and systemic risk events. We also provide a simulation algorithm for this process, which would be useful for statistical analysis, business applications and further research.