A0867
Title: Mean-field approximation of large banking network with defaults
Authors: Tomoyuki Ichiba - University of California Santa Barbara (United States) [presenting]
Abstract: The dynamics of the cash reserves of an interconnected banking system are considered. Whenever a bank defaults (by letting its reserve reach a given threshold), each bank is impacted negatively, via an instantaneous jump on its reserve level. We also take into account the arrival of new financial institutions in the system. The underlying dynamics of such system is written in the spirit of the spiking neural network models as studied previously. We study the mean field limit of such system and focus in particular on its stationary distribution and on learning about the network structure. We also discuss the effects of network structure in the mean-field system.
