TEI@I方法论及其在外汇汇率预测中的应用

TEI@I Methodology and Its Application to Exchange Rates Prediction

基于TE I@I方法论的理论框架,构建了一个基于TE I@I方法论的外汇汇率预测模型。在此模型中,传统的经济计量模型用于处理外汇汇率的主要趋势,人工神经网络技术用于分析外汇汇率的非线性,而文本挖掘和专家系统用于处理外汇市场中的突现性和不稳定性。最后,基于集成的思想,利用支持向量回归技术对上述3个部分进行非线性集成,从而获得一个更为精确的预测结果。通过实证方法验证了基于TE I@I方法论的外汇汇率预测模型的有效性。

On the basis of TEI@I methodology＇s theoretical framework, a TEI@I-based foreign exchange rates forecasting model is proposed, in which econometrical models are used to forecast the main trends of the rates, the nonlinear components of the rates are analyzed by using artificial neural network （ANN） models and the impacts of irregular and the infrequent future factors on the rates are explored using text mining and rule-based expert systems techniques. A fully novel nonlinear integrated forecasting approach with error correction and judgmental adjustment is formulated by means of support vector regression technique. For further illustration, the effectiveness of the TEI@I-based foreign exchange rates forecasting model was verified by the three foreign exchange rates.