摘要
针对中国沿海散货运价指数(CBFI)预测对精度的要求,从内在波动特性角度出发,提出一种基于集合经验模态分解(EEMD)-粒子群优化算法(PSO)-最小二乘法支持向量机(LSSVM)的组合预测模型.对比LSSVM、PSO-LSSVM、EMD-PSO-LSSVM三种预测模型,EEMD可对CBFI序列中波动较大数据进行降噪分解,保留序列的内在波动特性,且预测精度有一定提升,预测性能更佳.
To meet the requirement of China coastal bulk freight index(CBFI)prediction accuracy,from the perspective of internal fluctuation characteristics,a combined prediction model based on ensemble empirical mode decomposition(EEMD)-particle swarm optimization(PSO)-least squares support vector machine(LSSVM)was proposed.The comparison of three prediction models of LSSVM,PSO-LSSVM and EMD-PSO-LSSVM shows that EEMD can decompose and denoise the data with large fluctuation in CBFI sequence to reserve the inherent fluctuation characteristics of the sequence.Moreover,the prediction accuracy is improved to a certain extent,and the prediction performance is better.
作者
贾红雨
周晨昕
王宇涵
林岩
JIA Hong-yu;ZHU Chen-xin;WANG Yu-han;LIN Yan(Shipping Economics and Management College,Dalian Maritime University,Dalian 116026,China)
出处
《大连海事大学学报》
CAS
CSCD
北大核心
2020年第1期107-113,共7页
Journal of Dalian Maritime University
基金
国家自然科学基金面上项目(71571025).
