本研究旨在探討機器學習技術在台灣股票市場「底底高」策略中的應用。透過使用多種機器學習演算法，包括XGBoost、LightGBM和CatBoost，我們對2013年至2023年間的台股市場數據進行了分析和回測，並將結果與傳統技術分析方法進行比較。我們設計了一套嚴謹的回測框架，採用滑動窗口法動態調整訓練和驗證集，以提升模型的穩健性和預測準確性。研究過程中，我們引入了三個主要交易限制：每日成交量限制、閒置資金最大投資金額限制和單次最大投入金額限制，以分散投資風險和管理資金流動性。結果顯示，經過機器學習模型優化的交易策略在多個評估指標上均顯著優於傳統策略，表明機器學習在預測市場走勢和指導交易決策方面具有較高的準確性和穩定性。總結來說，本研究證明了機器學習技術在金融市場分析和交易策略優化中的潛力。透過應用這些先進模型，我們不僅提升了交易策略的預測能力，還有效控制了投資風險，實現了顯著的投資回報。這些研究成果為未來的投資策略設計和模型優化提供了寶貴的參考，並為實際應用中的資產管理和風險控制提供了新的思路和方法。

This study explores the use of machine learning techniques in the "Bottom-Up" strategy for the Taiwan stock market. We applied XGBoost, LightGBM, and CatBoost to analyze and backtest data from 2013 to 2023, comparing results with traditional technical analysis methods. A rigorous backtesting framework with a sliding window method was used to adjust training and validation sets dynamically.
We introduced trading constraints such as daily volume limits and maximum idle fund investment limit, and maximum single investment limit to manage risks. The machine learning-optimized strategies significantly outperformed traditional strategies in various metrics, showing high accuracy and stability in market trend prediction and trading decisions.
In summary, this study highlights the potential of machine learning in enhancing trading strategies and controlling investment risks, achieving notable returns. These findings provide valuable insights for future strategy design and practical asset management.

論文審定書 i
摘要 ii
Abstract iii
目錄 iv
圖次 v
表次 vi
第壹章 緒論 1
第一節 研究背景與動機 1
第二節 研究目的 3
第三節 研究流程 5
第貳章 文獻探討 6
第一節 技術分析理論與回顧 6
第二節 機器學習應用於股價趨勢 10
第參章 研究方法 14
第一節 研究架構 14
第二節 資料來源 17
第三節 交易策略設計 17
第四節 機器學習模型 17
第五節 機器學習特徵選用 23
第六節 機器學習時間設計 25
第肆章 實證結果與分析 28
第一節 初步策略回測與績效分析 28
第二節 滾動窗口法的應用 30
第三節 綜合資金回測 44
第伍章 結論 75
參考文獻 77

Appel, G. (2005). Technical Analysis: Power Tools for Active Investors.
Brock, W., Lakonishok, J., & LeBaron, B. (1992). A test of technical trading rules: A Monte Carlo experiment. Journal of Finance, 47(5), 1731-1764.
Bustos, O., & Pomares-Quimbaya, A. (2020). Stock market movement forecast: A systematic review. Expert Systems with Applications, 156, Article 113464.
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785-794). ACM.
Ellis, K., & Parbery, S. (2005). The effectiveness of technical analysis: The use of moving averages. Journal of Financial Markets, 8(2), 123-142.
Fischer, T., & Krauss, C. (2018). Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270(2), 654-669.
Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. Econometrica, 88(3), 1051-1100.
Heaton, J. B., Polson, N. G., & Witte, J. H. (2017). Deep learning in finance. Annual Review of Financial Economics, 9, 145-181.
Henrique, B. M., Sobreiro, V. A., & Kimura, H. (2019). Stock price prediction using support vector regression on daily and up to the minute prices. Journal of Finance and Data Science, 5(1), 12-18.
Huck, N. (2019). Machine learning for stock price prediction: Everything you need to know. Applied Soft Computing, 81, Article 105740.
Jegadeesh, N., & Titman, S. (1993). Returns to buying winners and selling losers: Implications for stock market efficiency. Journal of Finance, 48(1), 65-91.
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T. Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. In Advances in Neural Information Processing Systems (pp. 3146-3154).
Krauss, C., Do, X. A., & Huck, N. (2017). Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500. European Journal of Operational Research, 259(2), 689-702.
Kuang, L., Liu, Y., & Sun, Q. (2014). The application and profitability of technical analysis in emerging stock markets. Journal of Financial Markets, 21(4), 567-590.
Lo, A. W., Mamaysky, H., & Wang, J. (2000). Foundations of technical analysis: Computational algorithms, statistical inference, and empirical implementation. Journal of Finance, 55(4), 1705-1765.
Neely, C. J., Rapach, D. E., Tu, J., & Zhou, G. (2014). Forecasting the equity risk premium: The role of technical indicators. Management Science, 60(7), 1772-1791.
Nison, S. (1991). Japanese Candlestick Charting Techniques: A Contemporary Guide to the Ancient Investment Techniques of the Far East.
Osler, C., & Chang, K. (1995). Head and shoulders: Not just a flaky pattern. Financial Analysts Journal, 51(4), 23-34.
Park, C.-H., & Irwin, S. H. (2007). What do we know about the profitability of technical analysis? Journal of Economic Surveys, 21(4), 786-826.
Qiu, M., & Song, Y. (2016). Predicting the direction of stock market index movement using an optimized artificial neural network model. PLOS ONE, 11(5), e0155133.
Ticknor, J. L. (2013). A Bayesian regularized artificial neural network for stock market forecasting. Expert Systems with Applications, 40(14), 5501-5506.
Treynor, J. L., & Ferguson, R. (1985). In defense of technical analysis. Journal of Finance, 40(2), 757-773.
Zhou, Z., Dong, Y., & Wang, H. (2019). Stock market prediction on high-frequency data using generative adversarial nets. Mathematical Problems in Engineering, Article ID 7812525.