论文部分内容阅读
BDI指数的预测对于航运市场经营管理具有重要的指导作用。本研究构建了基于周期、均值回复和跳跃特性的BDI指数O-U随机预测模型,主要创新点有:一是分析了BDI指数的周期、均值回复以及跳跃特性,将该三大特性纳入BDI指数随机预测模型,有效提升了BDI指数预测理论科学性。二是借助O-U随机过程,建立了基于周期、均值回复以及跳跃特性BDI指数预测模型,同时,利用Fourier级数函数估计周期参数,借助一阶自回归估计均值回复参数,以及Gamma分布和双指数分布来估计跳跃参数,解决了参数较多、估计难度较大的问题。三是采集2013年-2015年BDI指数日数据进行拟合,并借助蒙特卡罗方法对2016年上半年BDI指数开展了预测,结果表明本模型预测精确度较高。
The forecast of BDI index plays an important guiding role in the management of shipping market. The main innovations of BDI index OU stochastic prediction model based on periodicity, average recovery and jump characteristics are as follows: First, the cycle, mean recovery and jump characteristics of BDI index are analyzed. These three characteristics are included in the BDI index stochastic prediction Model, effectively enhance the scientific index of BDI forecast theory. The second is to establish the BDI index forecasting model based on OU stochastic process with periodic, mean response and jump characteristics. At the same time, the Fourier series function is used to estimate the periodic parameters. The first order autoregressive method is used to estimate the mean response parameters, Gamma distribution and biexponential distribution To estimate the jump parameters, to solve the problem of more parameters and more difficult to estimate. Thirdly, the BDI index data of 2013-2015 were collected for fitting, and the BDI index of the first half of 2016 was predicted by means of Monte Carlo method. The result shows that the model predicts the accuracy is high.