基于豆图时间序列的高频金融交易分析  

Analysis of high frequency financial transactions based on beanplot time series

作　　者：袁晓惠 黄小峰 王晨 张晓蕊 刘元元 王宇婷 李佳彬 杨凯 YUAN Xiaohui;HUANG Xiaofeng;WANG Chen;ZHANG Xiaorui;LIU Yuanyuan;WANG Yuting;LI Jiabin;YANG Kai(School of Mathematics and Statistics,Changchun University of Technology,Changchun 130012,China)

机构地区：[1]长春工业大学数学与统计学院,吉林长春130012

出　　处：《长春工业大学学报》2024年第1期32-38,共7页Journal of Changchun University of Technology

基　　金：国家自然科学基金项目(11901053);吉林省自然科学基金项目(20220101038JC);吉林省大学生创新创业训练计划项目(S202210190126)。

摘　　要：基于豆图时间序列对贵州茅台股票的高频金融交易数据进行了重构,提取了每个豆图的均值,组成一个新的时间序列进行实证分析。在进行实证分析中,对比多个时间序列模型,最终选定ARIMA模型和TAR模型对成交价进行拟合,经过比较得到基于TAR模型进行预测更为合适。This paper reconstructs the high-frequency financial trading data of Moutai stock market based on the beanplot time series,extracts the average value of each beanplot,and forms a new time series for empirical analysis.In the empirical analysis,multiple time series models were compared,and two models,ARIMA model and TAR model,were ultimately selected to fit the transaction price.After comparison,it was found that predicting based on the TAR model is more suitable.

关 键 词：豆图时间序列 高频数据 ARIMA模型 TAR模型 预测 

分 类 号：O212[理学—概率论与数理统计]