Application of Neural Network Ensemble in NonlinearTime-Serials Forecasts
Sijun Peng, Siru Zhu · 2009
Neural network ensemble is developed as a new neural network model in recent years. It is a paradigm where a collection of a finite number of neural networks is trained for the same task. Compared with single neural network, ensemble model has significant improvement in the learning and generalization. This paper proposes the application of neural network ensemble in prediction for nonlinear time-serials. In numerical simulation, the Loreacutenz system's data are applied. The results show that ensemble network model has a good effect and it is suitable for the prediction of nonlinear time-serials.