Forecasting seasonal time series with computational intelligence: contribution of a combination of distinct methods.

Abstract

Accurate time series forecasting are important for displaying the manner in which the past continues to aect the future and for planning our day to day activities. In recent years, a large literature has evolved on the use of computational intelligence in many forecasting applications. In this paper, several computational intelligence techniques (genetic algorithms, neural networks, support vector machine, fuzzy rules) are combined in a distinct way to forecast a set of referenced time series. Forecasting performance is compared to the a standard and method frequently used in practice.