Hybrid Differencial Evolutionary System For Time Series Prediction

Title: Hybrid Differencial Evolutionary System For Time Series Prediction

Authors: Araújo, Ricardo de A.; Vasconcelos, Germano C.; Ferreira, Tiago A. E.

Resumo: This paper presents a Hybrid Differential Evolutionary System (HDES) for time series forecasting. It consists of an intelligent hybrid model composed of an Artificial Neural Network (ANN) combined with an Improved Differential Evolution (IDE). The IDE searches for the relevant time lags for a correct time series characterization, the number of processing units in the ANN hidden layer, the ANN training algorithm and the modeling of ANN. Initially, the proposed HDES chooses the most fitted forecasting model, thus it performs a behavioral statistical test in the attempt to adjust forecast time phase distortions that appear in some time series. An experimental analysis is conducted with the proposed HDES using three real world time series and five well-known performance measures are used to assess its performance. The obtained results are compared to MultiLayer Perceptron (MLP) networks and the previously introduced Time-delay Added Evolutionary Forecasting (TAEF) method.

Keywords: Artificial Neural Networks; Differential Evolution; Hybrid Systems; Time Series Forecasting

Pages: 6

DOI: 10.21528/CBRN2007-078

Paper as PDF: 50100078.pdf

BibTex file: 50100078.bib