Project overview
Network Connectivity has important implications in the risk management of economic, financial and societal phenomena such as financial contagion in stock markets, the transmission of risk spillovers and market exuberance as well as the monitoring of the effects of climate change. A popular modelling approach for capturing such interdependencies is to employ the Vector Autoregression Model with temporal and contemporaneous effects. On the other hand, more recently a different stream of literature has proposed the use of network-driven models. Complex networks can be modelled by such time series regressions (e.g., NVAR) which allow to incorporate features that capture both the underline network structure as well as node-specific characteristics (see, Zhu et al. (2017)). In this research project, we focus on the statistical modelling and asymptotic analysis of various network-driven econometric specifications that capture such time series dynamics under a more general dependence structure.