[1]邱凯星,伍文燕,陈 卓.基于组合 LSTM 的股票价格预测方法[J].计算机技术与发展,2022,32(S1):73-79.[doi:10. 3969 / j. issn. 1673-629X. 2022. S1. 017]
　QIU Kai-xing,WU Wen-yan,CHEN Zhuo.Stock Price Prediction Method Based on Combined-LSTM[J].,2022,32(S1):73-79.[doi:10. 3969 / j. issn. 1673-629X. 2022. S1. 017]
点击复制

基于组合 LSTM 的股票价格预测方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年S1期
页码:
73-79
栏目:
应用前沿与综合
出版日期:
2022-12-11

文章信息/Info

Title:
Stock Price Prediction Method Based on Combined-LSTM
文章编号:
1673-629X(2022)S1-0073-07
作者:
邱凯星1 ; 伍文燕1 ; 陈 卓2
1. 广东工业大学 计算机学院,广东 广州 510006;
2. 广东工业大学 自动化学院,广东 广州 510006
Author(s):
QIU Kai-xing1 ; WU Wen-yan1 ; CHEN Zhuo2
1. School of Computers,Guangdong University of Technology,Guangzhou 510006,China;
2. School of Automation,Guangdong University of Technology,Guangzhou 510006,China
关键词:
LSTM; 股价预测; 深度学习; 时间序列模型; 时间窗口
Keywords:
LSTM; stock price prediction; deep learning; time series model; time window
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2022. S1. 017
摘要:
在股票基金等二级投资市场上,投资人员往往依靠公司基本面分析、财报营利情况、结合 MACD 和 KDJ 等指标形成投资决策,难以把握非线性时间序列的股票价格走势。 针对以上情况,为提高传统时间序列模型因时间窗口选择所造成的预测精度受限等问题,提出一种基于组合 LSTM 的股票价格预测方法。 该方法使用多时间窗口综合预测的思路,可在多个不同时间跨度上学习股票的历史价格规律及趋势,从而进行有效的股票价格预测,相比于传统的非时间序列模型与单一时间序列模型具有更好的全局感受野与鲁棒性,有助于证券从业人员与个人投资者形成投资决策。 实验结果表明,本模型在 MSE、RMSE 和 MAE 三个误差指标上相比于普通的 LSTM 网络分别降低了 6. 80% 、3. 47% 和 4. 67% ,提升了股票价格预测的准确率并具有较好的普适性。
Abstract:
In secondary investment markets such as stock funds,stock investors often rely on company fundamental analysis, financialreporting and profitability,combined with MACD and KDJ indicators to form investment decisions,and it is difficult to grasp the trend ofstock prices in nonlinear time series. In view of the above situation,in order to improve the problem of limited prediction accuracy causedby the selection of time window of the traditional time series model, a stock price prediction method based on combined LSTM isproposed. This method uses the idea of multi - time window comprehensive forecasting,which can learn the historical price laws andtrends of stocks in multiple different time spans,so as to make effective stock price forecasts. Compared with the traditional non -timeseries model and single time series model,it has better overall receptive field and robustness,which is helpful for securities practitionersand individual investors to make investment decisions. The experiment shows that this model reduces 6. 80% ,3. 47% and 4. 67%respectively compared with the ordinary LSTM network in the three error indicators of MSE,RMSE and MAE. This model improves theaccuracy of stock price prediction and has good universality.

相似文献/References:

[1]张浩,张代远.基于三次样条权函数神经网络的股价预测[J].计算机技术与发展,2014,24(06):28.
　ZHANG Hao[],ZHANG Dai-yuan[].Stock Prediction Based on Neural Networks with Cubic Spline Weight Functions[J].,2014,24(S1):28.
[2]曹春萍,武 婷.多主题下基于 LSTM 语义关联的长文本过滤研究[J].计算机技术与发展,2019,29(11):1.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 001]
　CAO Chun-ping,WU Ting.Research on LSTM Semantic Correlation Long Text Filtering Based on Subject Dependence[J].,2019,29(S1):1.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 001]
[3]吴娟娟,任 帅,张卫钢,等.一种基于 LSTM 模型的日销售额预测方法[J].计算机技术与发展,2020,30(02):133.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 026]
　WU Juan-juan,REN Shuai,ZHANG Wei-gang,et al.A Daily Sales Forecasting Method Based on LSTM Model[J].,2020,30(S1):133.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 026]
[4]陈可心,黄 刚.CStock:一种结合新闻与股价的股票走势预测模型[J].计算机技术与发展,2020,30(09):18.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 004]
　CHEN Ke-xin,HUANG Gang.Cstock:A Stock Trend Forecasting Model Combining News and Stock Price[J].,2020,30(S1):18.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 004]
[5]高洁云,赵逢禹,刘 亚.基于语义增强的改进混合特征选择的文本分类[J].计算机技术与发展,2021,31(01):24.[doi:10. 3969 / j. issn. 1673-629X. 2021. 01. 005]
　GAO Jie-yun,ZHAO Feng-yu,LIU Ya.Text Classification of Modified Hybrid Feature Selection Based on Semantic Enhancement[J].,2021,31(S1):24.[doi:10. 3969 / j. issn. 1673-629X. 2021. 01. 005]
[6]王冰玉,刘勇军.基于模糊近似支持向量回归的股价预测研究[J].计算机技术与发展,2021,31(03):14.[doi:10. 3969 / j. issn. 1673-629X. 2021. 03. 003]
　WANG Bing-yu,LIU Yong-jun.Research on Stock Price Prediction Based on Fuzzy Proximal Support Vector Regression[J].,2021,31(S1):14.[doi:10. 3969 / j. issn. 1673-629X. 2021. 03. 003]
[7]勾志竟,宫志宏 *,刘布春.基于 TensorFlow 的 LSTM 算法在农业中的应用[J].计算机技术与发展,2021,31(08):215.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 037]
　GOU Zhi-jing,GONG Zhi-hong *,LIU Bu-chun.Application of LSTM in Rice Yield Prediction Based on TensorFlow[J].,2021,31(S1):215.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 037]
[8]黄文龙,薛未杰,冷佳俊,等.基于堆叠策略的高送转题材股预测模型[J].计算机技术与发展,2022,32(S1):89.[doi:10. 3969 / j. issn. 1673-629X. 2022. S1. 020]
　HUANG Wen-long,XUE Wei-jie,LENG Jia-jun,et al.Forecasting Model of High-Delivery-Rate Stock Based on Stacking Strategy[J].,2022,32(S1):89.[doi:10. 3969 / j. issn. 1673-629X. 2022. S1. 020]

更新日期/Last Update: 2022-06-10