基于集成分解的农产品价格预测
Research on agricultural commodity futures price prediction based on ensemble empirical mode decomposition
张博群 1孙倩 1沈虹1
作者信息
- 1. 扬州大学商学院,江苏 扬州 225127
- 折叠
摘要
深度学习在用于预测非线性时间序列时表现出色,且无须考虑变量之间的内生性问题.将集成经验模态分解(ensemble empirical mode decomposition,EEMD)方法与卷积神经网络(convolutional neural networks,CNN)、长短期记忆模型(long short-term memory,LSTM)、门控循环单元(gated recurrent units,GRU)相结合,构建基于集成分解的农产品期货价格预测模型.以中国玉米、棉花和大豆期货价格为例,对原始期货价格信号进行 EEMD分解,然后将分解向量分别输入深度学习模型中进行训练,最终得出 EEMD-GRU模型为最优价格预测模型.结果显示,与单独的深度学习模型相比,该文所提基于集成分解的组合模型在预测准确性方面优势明显,具有更强的泛化能力.
Abstract
Deep learning performs excellent in predicting nonlinear time series,and it does not con-sider the endogeneity between variables.This paper integrates the Ensemble Empirical Mode De-composition(EEMD)method with Convolutional Neural Network(CNN),Long Short-Term Memory(LSTM),and Gated Recurrent Unit(GRU),and a model for forecasting agricultural com-modity futures prices based on integrated decomposition is constructed.Taking Chinese corn,cot-ton and soybean futures prices as examples,the original futures price signal is decomposed by EE-MD,and then the decomposed vectors are input into the deep learning models for training.Finally,it is concluded that EEMD-GRU model is the optimal price prediction model.The results demon-strate that compared with the individual deep learning models,the proposed integrated EEMD mod-el has obvious advantages in predictive accuracy and stronger generalization ability.
关键词
农产品期货/集成经验模态分解/深度学习Key words
agricultural commodity futures/ensemble empirical mode decomposition(EEMD)/deep learning引用本文复制引用
基金项目
国家自然科学基金(61803331)
国家自然科学基金(92371116)
江苏省自然科学基金(BK20170515)
出版年
2024
NETL
NSTL
万方数据