研究石油价格优化预测网题,石油价格具有严重的非线性和因子难以精确确定,传统预测算法只能对石油价格线性关系进行,导致预测精度低.为提高石油价格预测精度,提出一种基于投影寻踪和神经网络算法的石油价格预测模型.首先根据相关研究选择石油价格影响因子,然后利用投影寻踪算法对影响因子进行筛选,将其得到因子作为BP神经网络输入变量节点,建立石油价格预测模型.仿真结果证明,改进算法能够很好地反映石油价格波动趋势,提高了石油价格的预测精度,为石油价格预测提供了一种高精度预测工具.%Oil price system has the characteristics of serious nonlinear and uncertainty, traditional algorithms can only treat linear relationship in oil price system, and the accuracy of prediction is low. In order to improve the predic-tion precision of oil price, the paper put forward an oil price forecast model based on projection pursuit and neural network algorithm. According to the related research, oil price influence factors were selected. Then projection pur-suit algorithm was used for impact factors screening to decide which factors were as the input variables of BP neural network, and finally, oil price forecasting model was built. The simulation results show that the algorithm can well reflect the fluctuation trends of oil prices, improve the accuracy of oil price, and provide a high accuracy prediction tool.
展开▼
