ABSTRACT
Modeling; of financial systems using neural network techniques has attracted a great deal of attention in the past few years. Neural networks, because of their inductive nature, can infer complex nonlinear relationships between input and output variables, and thus bypass the step of theory formulation. This paper reviews the state of the art in financial modeling using neural networks and describes applications in key areas, such as foreign exchange and fixed income. It shows that with careful network design, the backpropagation learning procedure is an effective way of training neural networks for time-series prediction.
