论文部分内容阅读
波罗的海运价指数一直受到航运与国际贸易界的高度关注,研究波罗的海运价指数的波动规律,探讨其发展变化趋势日益为人们所重视.当前对波罗的海运价指数的预测主要采用时间序列方法,这些方法计算过程复杂,而且适应性相对较弱.利用BP网络强大的非线性映射功能,建立了一个预测BFI走势的模型,分别从短期与中长期对波罗的海运价指数进行预测,并用实际数据进行了验证,结果表明,利用神经网络对BFI进行预测具有很高的预测精度.
The Baltic Freight Index has always been highly concerned by the shipping and international trade circles, studying the fluctuation law of the Baltic Freight Index and exploring the trend of its development and change. People pay more and more attention to the Baltic Freight Index. The current forecast of the Baltic Freight Index mainly adopts the time series method. The calculation process is complex and its adaptability is relatively weak.Based on the strong nonlinear mapping function of BP network, a model for predicting the trend of BFI is established and the Baltic Freight Index is predicted respectively in the short and medium term. The results show that using neural network to predict BFI has high prediction accuracy.