政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/136846
English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  Items with full text/Total items : 113822/144841 (79%)
Visitors : 51788571      Online Users : 563
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
    •  Adv. Search
    Home ‧ Login ‧ Upload ‧ Help ‧ About ‧ Administer Goto mobile version
    政大典藏 > College of Commerce > Department of MIS > Theses >  Item 140.119/136846


    Please use this identifier to cite or link to this item: https://nccur.lib.nccu.edu.tw/handle/140.119/136846


    Title: 基於深度學習+BERT與強化學習進行FAANG股價預測
    FAANG stock prediction based on Deep learning +BERT and Reinforcement learning
    Authors: 蔡政融
    Tsai, Cheng-Jung
    Contributors: 姜國輝
    Chiang, Kuo-Huie
    蔡政融
    Tsai, Cheng-Jung
    Keywords: 深度學習
    強化學習
    股票預測
    FAANG
    BERT
    Date: 2021
    Issue Date: 2021-09-02 15:54:38 (UTC+8)
    Abstract: 在本研究中，我們想對金融商品進行預測，並且透過投資報酬率來探討兩種深度學習的方法孰優孰劣，同時我們也認為除了股票價格和交易量之外，技術指標和新聞情緒都是影響股票走勢的重要因素之一，因此，在最後的結果中我們也會與大盤ETF (Exchange Traded Funds) GSPC(追蹤S&P 500) 和 QQQ(追蹤內斯達克指數)進行比較，用以衡量模型。
    首先我們從CNBC爬取五間公司(Facebook, Amazon, Apple, Netflix, Google)的新聞資料以及從yahoo股市中獲取2013年到2020年的股市資訊，再來使用BERT衡量新聞情緒，這裡將它定義為三種情緒(負面、中立、正面)，並透過加權平均獲得一天的情緒指標，為了有更多的特徵資料量，實驗中也加入技術指標如MACD ( Moving Average Directional Index ), RSI ( Relative Strength Index )等。接著，比較深度學習模型LSTM (Short Term Memory Networks)、GRU (Gated Recurrent Unit Network)和PPO (Proximal Policy Optimization)深度強化學習模型，在這五支股票中的表現。
    從本實驗中實證分析，可以得到以下結果 :
    從2018年6月25日到2020年12月31日期間，若直接投資S&P 500 指數，平均年化報酬率為15.56%，若是直接投資Dow Jones 指數，平均年化報酬率為9.39%，若是直接投資Nasdaq 指數，平均年化報酬率為27.94%，而透過直接持有FAANG，平均年化報酬率為24.39%。
    透過強化學習策略對上述四個標的投資平均年化報酬率為投資S&P 500 指數平均年化報酬率為28.05%，投資Dow Jones 平均年化報酬率為13.49%，投資Nasdaq 指數 平均年化報酬率為32.36%，投資FAANG 平均年化報酬率為25.57%。
    Reference: References
    1. the economics of big tech (2019) Wikipedia. Available at: https://zh.wikipedia.org/wiki/Bigtech
    2. the quantitative trading (2014) baike.baidu. Available at: https://baike.baidu.com/item/quantitativetrading/5266581
    3. the original of FinBert(2020)Available at: https://zhuanlan.zhihu.com/p/368795160
    4. William F. Sharpe (1994) : The Sharpe Ratio
    5. Filippo Petroni, Giulia Rotundo : Effectiveness of measures of performance during speculative bubbles (2007)
    6. Hadi S. Jomaa1, Josif Grabocka1, and Lars Schmidt-Thieme1 Hyp-RL : Hyperparameter Optimization by Reinforcement Learning (2019)
    7. Bergstra,J.,Bengio,Y.:Random search for hyper parameter optimization. Journal of Machine Learning Research 13(Feb), 281–305 (2012)
    8. Shahriari, B., Swersky, K., Wang, Z., Adams, R.P., De Freitas, N.: Taking the human out of the loop: A review of bayesian optimization. Proceedings of the IEEE 104(1), 148–175 (2016)
    9. Xu, Z., van Hasselt, H.P., Silver, D.: Meta-gradient reinforcement learning. In: Advances in Neural Information Processing Systems 31: Annual Conference on Neural Information Processing Systems 2018, NeurIPS 2018, 3-8 December 2018, Montr ́eal, Canada. pp. 2402–2413 (2018)
    10. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin: Attention Is All You Need (2017)
    11. Jacob Devlin, Ming-Wei Chang, Kenton Lee, Kristina Toutanova: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (2018)
    12. Zhuoran Xiong, Xiao-Yang Liu, Shan Zhong, Hongyang Yang, Anwar Walid: Practical Deep Reinforcement Learning Approach for Stock Trading(2018)
    Description: 碩士
    國立政治大學
    資訊管理學系
    108356023
    Source URI: http://thesis.lib.nccu.edu.tw/record/#G0108356023
    Data Type: thesis
    DOI: 10.6814/NCCU202101247
    Appears in Collections:[Department of MIS] Theses

    Files in This Item:

    File Description SizeFormat
    602301.pdf4616KbAdobe PDF20View/Open


    All items in 政大典藏 are protected by copyright, with all rights reserved.


    社群 sharing

    著作權政策宣告 Copyright Announcement
    1.本網站之數位內容為國立政治大學所收錄之機構典藏，無償提供學術研究與公眾教育等公益性使用，惟仍請適度，合理使用本網站之內容，以尊重著作權人之權益。商業上之利用，則請先取得著作權人之授權。
    The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

    2.本網站之製作，已盡力防止侵害著作權人之權益，如仍發現本網站之數位內容有侵害著作權人權益情事者，請權利人通知本網站維護人員(nccur@nccu.edu.tw)，維護人員將立即採取移除該數位著作等補救措施。
    NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback