Contribution of the neural networks in the forecasting of financial markets with memory

Abstract

The method of the neural networks is beginning to have an impressive range of applications in various fields. Faced with the hope attached by these neural methods by piercing the mysteries of the stock market, appears the notion of long memory and his "conflict" with the efficient market hypothesis. In this article we will try to see the contribution made by this method to beat the old prediction methods such as that of technical analysis in forecasting the financial markets different in terms of their economic development.