摘要
运用经验模态分解(EMD)、人工神经网络(ANN)和时间序列,基于分解—重构—集成的思想,构建了一个组合预测模型。在模型的构建过程中,提出了对股票指数序列进行逐日前向滚动EMD分解的思路,将分解后的本征模函数(IMF)分量输入神经网络进行组合预测。运用上述基于前向滚动EMD模型分析沪深300指数和澳大利亚指数的波动特点和走势。结果显示:前向滚动EMD模型比ARIMA模型、GARCH模型和BP神经网络模型具有更高的预测精度。
A new combined forecasting model is built in this paper by using empirical mode decomposition(EMD), artificial neural network (ANN) and time series methods based on the idea of decomposition-reconstruction-integration. During the process of building this model,a new idea to decompose the stock index sequence by forward rolling EMD method is proposed. After decomposition,the intrinsic mode function(IMF) components are input into neural network to implement the combination forecast. Then this model is used to analyze the fluctuation characteris tics and the trend of Chinese Stock Index(CSI300) and Australian stock index. Empirical analysis result shows that, comparing with ARIMA model,GARCH model and BP neural network model,forward rolling EMD model obtains better forecasting result.
出处
《技术经济》
CSSCI
北大核心
2015年第5期70-77,共8页
Journal of Technology Economics
基金
中国智能金融研究院"融市场预测模型项目"(2014-2016)