Time series forecasting using Artificial Neural Networks vs. evolving models

Time series forecasting plays an important role in many fields such as economics, finance, business intelligence, natural sciences, and the social sciences. This forecasting task can be achieved by using different techniques such as statistical methods or Artificial Neural Networks (ANN). In this paper, we present two different approaches to time series forecasting: evolving Takagi-Sugeno (eTS) fuzzy model and ANN. These two different methods will be compared taking into account the different characteristic of each approach.