Application of Artificial Intelligence Methods to Prediction of Financial Time Series

Beside the methods based on classical statistic and econometrics, some artificial intelligence (AI) based approaches can be applied to predict the future development of financial time series data. In this paper some AI based algorithms which come from the area of Machine Learning (ML) and Neural Nets (NN) will be introduced and their practicability to predict the development of financial time series will be demonstrated. To make prediction with ML algorithms the regression tree algorithm is used. This algorithm can be seen as a modified form of Quinlan’s ID3 algorithms (see [2]). It is described in detail by Breiman et al in their book about the CART (classification and regression trees) algorithm (see [1]). Using NN for prediction, a backpropagation algorithm is used (see [4]).