Development of an automated trading system to the stock exchange and forex using neural networks multilayer perceptron

Abstract

The task of trying to predict financial market behavior has always attracted the interest of traders, as their assumptions may result in profits from the investment made. The financial market is strongly influenced by the information that comes from economic and political situation of a country to natural disasters. With the popularization of the internet, the volume of information that is available influences the market, making it more difficult to predict the impact this information may have, including for experienced investors. Due to this difficulty, many studies are being carried out associating technical analysis and computational intelligence approaches such as neural networks to try to predict market behavior. Thus, the negotiations are influenced by historical data and decisions are not made on the basis of investor emotional or psychological state.