Ensemble Recurrent Neural Networks for Complex Time Series Prediction with Integration Methods.

In this paper the development of the general architecture of a method to design recurrent ensemble neural networks for time series prediction is presented. Therefore, experiments are shown for the ensemble recurrent neural network, as well as the integration of the responses of the ensemble recurrent neural network, with integration by average, weighted average integration, type-1, type-2 and Generalized Type-2 fuzzy systems. The time series used to test the proposed architecture is that of petroleum. The simulation results show the effectiveness of the proposed method.