Visual monitoring of financial stability with a self-organizing neural network

Since the outset of the deregulation of international financial markets in the 1980s, the frequency of currency crises has increased. Solely in the 1990s, five global storms of financial turmoil, also including collapses of the currency, have occurred. To date, crisis forecasting and monitoring of financial stability is still at a preliminary stage. This paper explores whether the application of the Self-Organizing Map (SOM), a neural network-based visualization tool, facilitates the monitoring of multidimensional economic data. The paper presents a visualization of both the evolution of economic indicators over time and of benchmarking countries, on a given point in time, as to their vulnerability for an imminent crisis. The results of this paper indicate that the SOM is a feasible tool for dynamic visualization of currency crises' early warning signals.