English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  全文筆數/總筆數 : 118405/149442 (79%)
造訪人次 : 78285388      線上人數 : 18
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 查詢小技巧：
  • 您可在西文檢索詞彙前後加上"雙引號"，以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋，建議至進階搜尋限定作者欄位，可獲得較完整資料
    •  進階搜尋


    請使用永久網址來引用或連結此文件: https://nccur.lib.nccu.edu.tw/handle/140.119/159233


    題名: 端對端創新模型：台灣股市投資組合建構與波動度管理之應用
    Application of Innovation End-to-End Model to Portfolio Construction and Volatility Management in the Taiwan Stock Market
    作者: 楊煥濬
    Yang, Huan-Chun
    貢獻者: 黃泓智
    楊煥濬
    Yang, Huan-Chun
    關鍵詞: 機器學習
    End-to-End模型
    波動度管理
    Machine Learning
    End-to-End model
    Volatility Management
    日期: 2025
    上傳時間: 2025-09-01 16:03:00 (UTC+8)
    摘要: 本研究探討台灣股票市場，使用結合預測與配置之End-to-End的模型，以減少傳統預測再配置之兩階段方法容易產生預測誤差累積，導致投組績效不穩定之問題。本文除使用CNN-LSTM模型以更好捕捉特徵與時間序列關係外，亦改良模型結構與損失函數、引入具溫度係數之softmax函數用以配置與模型訓練，並以投組夏普指數最大化為目標。藉由此改良方式，除可使模型用於較多數量之標的，更可同時達成選股與配置之任務。經回測實證，使用End-to-End模型所組成之投資組合，其夏普指數與最大回落，普遍優於傳統兩階段方法，而模型預測出之權重亦對於投組績效有正向影響，顯示了本研究使用之End-to-End模型兼具泛用性、有效性與穩健性。
    除證明End-to-End模型較傳統方法表現優異外。本研究納入較低波動之ETF進行波動度管理，並依照持有標的種類之數量，區分三種相異投資客群。並探討：固定比例、目標波動度與波動度上限三種方法，於各客群最佳之管理方式是否有差異，並為每種客群找尋最佳策略。
    實證結果顯示，納入低波動標的，對所有客群皆有助於提升夏普指數。而隨著持有種類數目增加，可將股票佔比隨之提高。不同客群依照目標波動度法進行計算與配置時，小型客群用以估算之計算天數應可較短，隨規模增加逐漸增至120天左右，可同時獲得較高之夏普指數並控制投組波動。波動度上限部分，探討使用不同監測天數窗口與不同監測門檻於不同投資規模客群之效果。其結果顯示，小型客群可用投組中短期之波動做為觸發條件；大型客群則可設立短期內大幅波動做為觸發條件。觸發次數過於頻繁，除了實務上管理不便外，也容易使交易成本過高而降低投資收益。因此本文最後比較固定監測天數與固定監測指標門檻兩種波動度管理方式。分析回測結果，兩種方式皆可達成相似結果，然而使用固定監測門檻之方法，其觸發次數較穩定。
    This study explores the Taiwan stock market using an end-to-end model that integrates prediction and portfolio allocation. This approach aims to mitigate performance instability caused by error accumulation in traditional two-stage methods. The model utilizes a CNN-LSTM architecture, with improved structure and loss function, and incorporates a temperature-based softmax function to maximize the portfolio's Sharpe ratio. This allows for simultaneous stock selection and allocation across numerous assets.
    Results shows the end-to-end model yields superior Sharpe ratios and maximum drawdowns compared to traditional methods, demonstrating its versatility, effectiveness, and robustness.
    The research also incorporates low-volatility ETFs for risk management and segments investors by portfolio size. It examines three strategies: fixed ratio, target volatility, and volatility cap. Results indicate that including ETFs improves Sharpe ratios across all investor groups. Larger portfolios benefit from higher stock allocations and longer lookback windows. For volatility caps, small investors should use short-to-mid-term volatility as triggers, while large investors respond to sudden short-term fluctuations. Overly frequent triggering, in addition to posing practical management inconveniences, also tends to inflate transaction costs, thus eroding investment returns. Finally, we compare fixed-window and fixed-threshold monitoring, the latter provides more stable adjustment signals.
    參考文獻: [1] Afan Galih Salman, Y., Heryadi, Y., Abdurahman, E., & Suparta, W. (2018). Single Layer & Multi-layer Long Short-Term Memory (LSTM) Model with Intermediate Variables for Weather Forecasting. Procedia Computer Science, 135, 89–98. https://doi.org/10.1016/j.procs.2018.08.153
    [2] Chevalier, G., Coqueret, G., & Raffinot, T. (2022). Supervised portfolios. Quantitative Finance, 22(12), 2275–2295. https://doi.org/10.1080/14697688.2022.2122543
    [3] Connell, P., & Hodgson, M. (2016). Managing investment outcomes with volatility control. Schroder Investment Management North America Inc.
    [4] DeMiguel, V., Garlappi, L., & Uppal, R. (2009). Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy? The Review of Financial Studies, 22(5), 1915–1953. https://doi.org/10.1093/rfs/hhm075
    [5] Hinton, G., Vinyals, O., & Dean, J. (2015). Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531. https://doi.org/10.48550/arXiv.1503.02531
    [6] Hochreiter, S., & Schmidhuber, J. (1997). Long Short-term Memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
    [7] Hsu, J. C., & Li, F. (2013). Low-Volatility Investing. Journal of Index Investing, 4(2). Available at SSRN: https://ssrn.com/abstract=2913306
    [8] LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278–2324. https://doi.org/10.1109/5.726791
    [9] Livieris, I. E., Pintelas, E., & Pintelas, P. (2020). A CNN–LSTM model for gold price time-series forecasting. Neural Computing and Applications, 32(23), 17351–17360. https://doi.org/10.1007/s00521-020-04867-x
    [10] Lu, W., Li, J., Li, Y., Sun, A., & Wang, J. (2020). A CNN-LSTM-based model to forecast stock prices. Complexity, 2020, Article 6622927. https://doi.org/10.1155/2020/6622927
    [11] Markowitz, H. (1952). Portfolio selection. The Journal of Finance, 7(1), 77–91.
    [12] Mei, X., Wang, Y., & Zhu, W. (2023). Bayesian nonparametric portfolio selection with rolling maximum drawdown control. Quantitative Finance, 23(10), 1497–1510. https://doi.org/10.1080/14697688.2023.2250386
    [13] Nguyen, K.-B., & Park, C. J. (2025). On Calibration of Prompt Learning Using Temperature Scaling. IEEE Access, 13, 31171–31182. https://doi.org/10.1109/ACCESS.2025.353861
    [14] Nystrup, P., Boyd, S., Lindström, E., & Madsen, H. (2019). Multi-period portfolio selection with drawdown control. Annals of Operations Research, 282, 245–271. https://doi.org/10.1007/s10479-018-2947-3
    [15] Tamplin, T. (n.d.). Equal-weighted portfolio: Definition, pros & cons, and factors. Finance Strategists. https://www.financestrategists.com/wealth-management/risk-profile/diversification/equal-weighted-portfolio/
    [16] Uysal, A. S., Li, X., & Mulvey, J. M. (2024). End-to-End risk budgeting portfolio optimization with neural networks. Annals of Operations Research, 339, 397–426. https://doi.org/10.1007/s10479-023-05539-4
    [17] Yadav, A., Jha, C. K., & Sharan, A. (2020). Optimizing LSTM for time series prediction in Indian stock market. Procedia Computer Science, 167, 2091–2100. https://doi.org/10.1016/j.procs.2020.03.257
    [18] Zhang, C., Zhang, Z., Cucuringu, M., & Zohren, S. (2021). A Universal End-to-End Approach to Portfolio Optimization via Deep Learning. arXiv preprint arXiv:2111.09170.https://doi.org/10.48550/arXiv.2111.09170
    [19] Zhang, Z., Zohren, S., & Roberts, S. (2020). Deep learning for portfolio optimization. The Journal of Financial Data Science, 2(4), 8–20. https://doi.org/10.3905/jfds.2020.1.042
    描述: 碩士
    國立政治大學
    風險管理與保險學系
    112358016
    資料來源: http://thesis.lib.nccu.edu.tw/record/#G0112358016
    資料類型: thesis
    顯示於類別:[風險管理與保險學系] 學位論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    801601.pdf1579KbAdobe PDF0檢視/開啟


    在政大典藏中所有的資料項目都受到原著作權保護.


    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 ©   - 回饋