A FRACTAL STRUCTURE OF THE TIME SERIES AND PREDICTION (A COMPARATIVE STUDY TO NEURAL NETWORK)

In this paper we develop two models for analysis and forecasting of financial time series. The first model is based on Elliott waves, which disposes of fractal structure. The second one uses an artificial neural network that is adapted by backpropagation. The Elliott wave principle is a detailed description of how financial markets behave and its analysis is a form of technical analysis and of behavioral economics that attempts to forecast trends by identifying extremes in investor psychology, tops and bottoms in markets, and other collective activities. Artificial neural networks are suitable for predicting time series mainly because of learning only from examples, without any need to add additional information that can bring more confusion than prediction effect. Neural networks are able to generalize and are resistant to noise. On the other hand, it is generally not possible to determine exactly what a neural network learned and it is also hard to estimate possible prediction error. They are ideal especially when we do not have any other description of the observed series. This paper also includes experimental results of time series prediction carried out with both mentioned models.