Time series prediction with neural networks for the Athens Stock Exchange indicator

Abstract

The main aim of this study is to predict the daily stock exchange price index of the Athens Stock Exchange (ASE) using back propagation neural networks. We construct the neural network based on the minimum embedding dimension of the corresponding strange attractor. Multistep prediction for nine days ahead is achieved with this particular network indicating the increased possibility of this technique for immediate forecasts for very timeshort data sets, mostly daily and weekly.