Insurance Risk in Networks

We plan to consider networks of financial firms that cross-hold each other and are vulnerability to external shock. We are interested in different types of external shocks including, but not limited to, market crises and cyber-attacks.

Once an external shock has occurred, we will model its spread throughout the financial firms' network as one of well-known in applied probability processes, for instance Susceptible-Infected-Susceptible (SIS) or Susceptible-Infected-Recovered-Susceptible (SIRS). We plan to use Markovian as well as non-Markovian processes.

When such shocks occur in financial networks, we intend to investigate the overall losses suffered with a particular focus on the emergence of large clusters of affected firms. We will consider how the network topology impacts the probability of the emergence of such clusters, and the distribution of the overall losses. We will study the effect of both the size of the network, as well as some measures on its structure (degree sequence, connectedness, presence of communities, etc).