The number of financial and economic transactions happening every second is ever-increasing, occurring on a variety of marketplaces, performed by different market players, and related to more and more interdependent complex instruments, services, and products. Ensuring financial stability and monitoring the global economy requires making sense of multiple data structures and data sources. Stochastic modelling remains a fundamental benchmark in finance, but complex market behaviours require data-driven modelling hardly captured by standard linear and univariate models. In this talk, we will introduce a series of methodologies, such as dynamic network models, non-homogeneous Poisson processes and various neural network architectures, used to capture dynamic and non-linear interdependencies between transactions and to provide timely and informative indicators about economic and financial systems.
