基于动态递归神经网络模型的混沌时间序列预测

提出了一种动态递归神经网络模型进行混沌时间序列预测，以最佳延迟时间为间隔的最小嵌入维数作为递归神经网络的输入维数，并按预测相点步进动态递归的生成训练数据，利用混沌特性处理样本及优化网络结构，用递归神经网络映射混沌相空间相点演化的非线性关系，提高了预测精度和稳定性。将该模型应用于Lorenz系统数据仿真以及沪市股票综合指数预测，其结果与已有网络模型预测的结果相比较，精度有很大提高。因此，证明了该预测模型在实际混沌时间序列预测领域的有效性和实用性。

Chaotic time series forecasting based on dynamic recurrent neural networks model

A Dynamic Recurrent Neural Networks(DRNN) model was presented in this paper to forecast chaotic time series.The input dimension of DRNN was decided by minimal embedding dimension.The training samples were generated by means of the stepping recursive phase points.It can improve precision and stability of prediction to use chaotic characteristic to deal with samples and mapping nonlinear function by DRNN.The DRNN model was applied to simulation of Lorenz system and shot-term forecasting of Shanghai stock index.Compared with the traditional standard BP neural network,this proposed model shows higher precision.Therefore,this research proves the effectiveness of the proposed model in the practical prediction of time series.