摘要
提升时间序列模型的预测精度需要全面了解其数据的线性和非线性复合特征,利用ARIMA以及RNN模型分别对时间序列进行建模,挖掘其线性以及非线性规律,最后得到两种模型的综合预估结果.选取沪深300指数(000300)2006年1月4日~2021年11月26日中所有交易日的K线数据为样本,分析结果说明,ARIMA-RNN混合模型的精度比单一循环神经网络模型的预测精度要高,混合模型对于短期动态与静态预测成效较高,有利于投资者和企业做出更加科学可行的决策.
Improving the prediction accuracy of a time series model required a comprehensive understanding of the linear and nonlinear composite characteristics of its data,and this paper used ARIMA and RNN models to model the time series respectively,and mined the linear and nonlinear laws,and finally obtained the comprehensive prediction results of the two models.This paper selected the K-line data of all trading days of the CSI 300 Index(000300)from January 4,2006 to November 26,2021 as the sample,and the analysis results showed that the accuracy of the ARIMA-RNN hybrid model was higher than that of the single recurrent neural network model,and the hybrid model had a higher effect on short-term dynamic and static prediction,which was conducive to investors and enterprises to make more scientific and feasible decisions.
作者
管学英
GUAN Xueying(School of Mathematics and Statistics,Chongqing Technology and Business University,Chongqing 400067,China)
出处
《哈尔滨商业大学学报（自然科学版）》
CAS
2024年第2期250-256,共7页
Journal of Harbin University of Commerce：Natural Sciences Edition
基金
经济社会应用统计重庆市重点实验室重点项目(KFJJ2022056)。
