One curious little fact about the Riemann zeta function is that if you evaluate its derivatives at the zeros of zeta, then on average this is real and positive (even though the function is complex). This has been proven for some time now, but the aim of this talk is to generalise the question further (higher derivatives, complex moments) and gain insight using random matrix theory. The takeaway message will be that there are a multitude of different proof techniques in RMT, each with their own advantages

In this talk, I will introduce a dynamic model of a banking network in which the value of interbank obligations is continuously adjusted to reflect counterparty default risk. An interesting feature of the model is that the credit value adjustments increase volatility during downturns, leading to endogenous distress contagion. The counterparty default risk can be computed backwards in time from the obligations' maturity date, leading to a specification of the model in terms of a forward-backward stochastic differential equation (FBSDE), coupled through the banks' default times. The singular nature of this coupling, makes a probabilistic analysis of the FBSDE challenging. So, instead, we derive a characterisation of the default probabilities through a cascade of partial differential equations (PDE). Each PDE represents a configuration with a different number of defaulted banks and has a free boundary that coincides with the banks' default thresholds. We establish classical well-posedness of this PDE cascade, from which we derive existence and uniqueness of the FBSDE.