Overview

Measures from information theory, such as Shannon entropy, have been used in a variety of economic and financial applications. In this work, we develop a data-driven econometric framework that promotes an understanding of the relationship between financial institutes using directed information graphs (DIGs). DIG is an information-theoretical generalization of Granger-causality which is suitable for capturing the interconnections between a set of high-dimensional time series with linear or nonlinear dynamics. We discuss how the structure of financial systems can be represented as a complex network, where the nodes are financial institutes and the connections are formed utilizing a connectivity measure. A weighted directed network can then be formed to identify the level of systemic risk in the financial sector. We also show (theoretically and through the simulation) how the proposed framework improves the measurement of systemic risk and explain its link to several well-known econometric models such as vector autoregression, switching models, and others.