摘要
价格数据数值与趋势的准确预测一直是金融风险量化控制的一大难题。在国际油价受外部因素影响剧烈波动的背景下,针对航空燃油价格预测问题,提出一种基于经验模态分解(EMD)和自回归滑动平均模型(ARMA)的非线性混合预测方法。研究结果表明,EMD-ARMA组合模型对非平稳时间序列信号的预测有效,精度相比较单一的ARMA模型有显著提高。
The accurate prediction of price data values and trends has always been a major problem in the quantitative control of financial risks.In the context of the international oil price being fluctuated by external influences in recent years, a nonlinear hybrid prediction method based on empirical mode decomposition(EMD) and autoregressive moving average model(ARMA) is proposed for the prediction of aviation kerosene price. The experimental results of the paper show that the EMD-ARMA combined model is effective for predicting non-stationary time series signals, and the accuracy is significantly improved compared with the single ARMA model.
作者
高伦
张心成
GAO Lun;ZHANG Xin-cheng(VSB-Technical University of Ostrava,Ostrava 70200,Czech Republic;Zhongnan University of Economics and Law,Wuhan 430073,China)
出处
《金陵科技学院学报（社会科学版）》
2019年第4期15-20,共6页
Journal of Jinling Institute of Technology(Social Sciences Edition)
基金
江苏省望云智库省级项目“关于国际贸易战略研究”(2019SHJ86)
关键词
组合预测
经验模态分解
ARMA
聚类
过度分解
combined forecasting
empirical mode decomposition(EMD)
ARMA
clustering
excessive decomposition
作者简介
高伦(1995-),男,江苏淮安人,博士研究生,主要从事公司风险量化、期权定价、数量经济研究。
