NEURAL NETWORKS, GENETIC ALGORITHMS AND STOCK TRADING

In this paper we present an application of neural networks and genetic algorithms to the analysis of the stock markets dynamics. The aim of this research is to find the optimal strategy for a stock trader whose decision system is made of a feed-forward, three-layer sequential neural network. The search of the optimal network is done using genetic algorithms. The main aspect developed in this paper is the determination of the number of hidden neurons. Numerical simulations has been made on the daily prices of the FIAT share and some result are reported.