Novel SVM-based neural network ensemble model for foreign exchange rates forecasting
In this study, a triple-phase support vector machine based neural network ensemble model is proposed for exchange rates forecasting. In the first phase, many different single neural network models are generated. In the second phase, a conditional generalized variance minimization method is used to select the appropriate ensemble members. In the final phase, the support vector machine regression method is used for neural network ensemble for prediction purpose. For further illustration, two exchange rate series are used for testing. Empirical results obtained reveal that this novel neural network ensemble model can improve the performance of foreign exchange rates forecasting.