[1]许飞飞,胡 月,汪召兵.基于生成式对抗网络的股价预测研究[J].浙江科技大学学报,2022,(03):207-215.[doi:10.3969/j.issn.1671-8798.2022.03.002 ]
　XU Feifei,HU Yue,WANG Zhaobing.Research on stock price forecasting based on FWGAN model[J].,2022,(03):207-215.[doi:10.3969/j.issn.1671-8798.2022.03.002 ]
点击复制

基于生成式对抗网络的股价预测研究(/HTML)
分享到：

《浙江科技大学学报》[ISSN:2097-5236/CN:33-1431/Z]

卷:
期数:
2022年03期
页码:
207-215
栏目:
出版日期:
2022-06-30

文章信息/Info

Title:
Research on stock price forecasting based on FWGAN model
文章编号:
1671-8798(2022)03-0207-09
作者:
许飞飞; 胡 月; 汪召兵
(浙江科技学院 理学院,杭州 310023)
Author(s):
XU Feifei;  HU Yue;  WANG Zhaobing
(School of Sciences, Zhejiang University of Science and Technology, Hangzhou 310023, Zhejiang, China)
关键词:
股价预测;  股民情绪;  生成式对抗网络股价预测模型;  时间序列
分类号:
F830.91
DOI:
10.3969/j.issn.1671-8798.2022.03.002
文献标志码:
A
摘要:
为了降低股票市场中噪声信息和投资者情绪对股票价格的影响,以便给投资者带来较高的投资回报并降低交易风险,特提出一种基于金融双向编码器表征和瓦瑟斯坦距离的生成式对抗网络(financial bidirectional encoder representation from transformers and Wasserstein generative adversarial networks,FWGAN)股价预测模型。本模型首先采集东方财富网股评数据,并利用自然语言处理预训练模型将股评数据量化为情绪值,然后将情绪值连同历史股票交易数据、技术指标数据输入由长短期记忆网络(long-short-term memory,LSTM)为生成器和卷积神经网络(convolutional neural network,CNN)为判别器组成的FWGAN模型中进行训练。对比LSTM模型、门控神经网络(gated recurrent units,GRU)模型和生成式对抗网络(generative adversarial networks,GAN)模型对山西汾酒股价的预测性能,结果表明,FWGAN模型的均方根误差为2.572,达到最低,预测效果最好。试验结果验证了本模型对股票时间序列预测的有效性和优越性,可以为投资者进行股价预测提供参考。

参考文献/References:

[1] 张晨希,张燕平,张迎春,等.基于支持向量机的股票预测[J].计算机技术与发展,2006(6):35.
[2] AKITA R, YOSHIHARA A, MATSUBARA T,et al. Deep learning for stock prediction using numerical and textual information[EB/OL].(2016-07-26)[2021-03-23].https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7550882.
[3] 彭燕,刘宇红,张荣芬.基于LSTM的股票价格预测建模与分析[J].计算机工程与应用,2019,55(11):209.
[4] 史建楠,邹俊忠,张见,等.基于DMD -LSTM模型的股票价格时间序列预测研究[J].计算机应用研究,2020,37(3):662.
[5] 赵红蕊,薛雷.基于LSTM-CNN-CBAM模型的股票预测研究[J].计算机工程与应用,2021,57(3):203.
[6] RAHMAN A S A, ABDUL-RAHMAN S, MUTALIB S. Mining textual terms for stock market prediction analysis using financial news[C]//International Conference on Soft Computing in Data Science. Singapore:Springer,2017.
[7] ZHANG X, ZHANG Y J, WANG S Z, et al. Improving stock market prediction via heterogeneous information fusion [J].Knowledge-Based Systems,2018,143:236.
[8] 张明禄.新闻事件驱动的市场预测研究[D].上海:上海交通大学,2013.
[9] SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the inception architecture for computer vision[C]//Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE,2016:2818.
[10] HENSMAN P, AIZAWA K. cGAN-based manga colorization using a single training image[EB/OL].(2017-05-30)[2021-03-23].https://arxiv.org/pdf/1706.06918.pdf.
[11] 罗会兰,敖阳,袁璞.一种生成对抗网络用于图像修复的方法[J].电子学报,2020,48(10):1891.
[12] ZHOU X Y, PAN Z S, HU G Y, et al. Stock market prediction on high-frequency data using generative adversarial nets[J].Mathematical Problems in Engineering,2018,special issue:1.
[13] ZHANG K, ZHONG G Q, DONG J Y, et al. Stock market prediction based on generative adversarial network[J].Procedia Computer Science,2019,147:400.
[14] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J].Neural Computation,1997,9(8):1735.
[15] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J].Advances in Neural Information Processing Systems,2014,3:2672.
[16] GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved training of Wasserstein GANs[C]//Proceedings of the 31st Annual Conference on Neural Information Processing Systems. New York:Neural Information Processing Systems Foundation,2017:5769.
[17] 李舟军,范宇,吴贤杰.面向自然语言处理的预训练技术研究综述[J].计算机科学,2020,47(3):170.
[18] 王子牛,姜猛,高建瓴,等.基于BERT的中文命名实体识别方法[J].计算机科学,2019,46(增刊2):138.
[19] CORREIA G M, NICULAE V, MARTINS A F T. Adaptively sparse transformers[EB/OL].(2019-08-30)[2021-03-23].https://arxiv.org/pdf/1909.00015v1.pdf.
[20] ARACI D T. FinBERT:financial sentiment analysis with pre-trained language models[EB/OL].(2019-06-25)[2021-03-23].https://arxiv.org/pdf/1908.10063.pdf.
[21] 侍冰雪,魏慧茹,朱韶东,等.基于网络结构的股票相关性研究[J].浙江科技学院学报,2015,27(1):62.
[22] 张杰,王茁.基于傅里叶分析法的股票市场超高频数据相关性分析[J].数学的实践与认识,2010,40(7):63.

相似文献/References:

[1]付乐,胡月,董虹伶,等.多时间尺度下变体生成式对抗网络的股价预测[J].浙江科技大学学报,2023,(01):72.
　FU Le,HU Yue,DONG Hongling,et al.Stock price prediction with a variant generative adversarial network in multiple time scales[J].,2023,(03):72.

备注/Memo

备注/Memo:
收稿日期:2021-03-26
基金项目:浙江省科技计划项目(2015C33088)
通信作者:胡 月(1964— ),男,河南省西峡人,教授,硕士,主要从事概率论极限理论和金融数学研究。E-mail:huyue@zust.edu.cn。
更新日期/Last Update: 2022-06-30