论文部分内容阅读
提出建立多维泰勒网动力学模型及参数辨识方法,和基于小波多维泰勒网模型的金融时间序列预测方法.利用Mallat算法将金融时间序列分解成一个低频信号和若干个高频信号;对不同频率的时间序列建立多维泰勒网动力学模型;通过共轭梯度法训练模型参数,并进行预测;将各模型的预测结果进行叠加,得到原始序列的预测值.实验结果表明,这种金融时间序列预测方法具有较高的预测精度和预测方向正确率.
This paper proposes the establishment of multi-dimensional Taylor network dynamics model and parameter identification method, and the prediction method of financial time series based on the wavelet multi-dimensional Taylor network model. The Mallat algorithm is used to decompose the financial time series into a low frequency signal and a number of high frequency signals; Time series to build multi-dimensional Taylor network dynamic model, the model parameters are trained by the conjugate gradient method, and the prediction is carried out, and the prediction results of each model are superposed to obtain the predicted value of the original sequence.The experimental results show that this method of forecasting financial time series Has a high prediction accuracy and prediction direction accuracy.