Control of chaotic finance system using artificial neural networks

Abstract

In this paper, the control of a chaotic finance system is applied by using Artificial Neural Networks (ANNs). Economic systems become more complicated and have undesired nonlinear factors. It is difficult to control when a chaotic behaviour occurs. So that the ANNs have the ability of learning functional relations, they can achieve the control of chaotic systems more effectively. On-line neural training algorithms are used for regulating the chaotic finance system to its equilibrium points in the state space. For faster training in back-propagation. Levenberg-Marquardt algorithm is preferred. Numerical simulations are performed to demonstrate the effectiveness of the proposed control technique.