基於隨機森林與 XGBoost 之加密貨幣短線交易訊號預測

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：20

、訪客IP：18.97.14.91

姓名

李卓儀(Cho-I Lee) 查詢紙本館藏

畢業系所

資訊管理學系在職專班

論文名稱

基於隨機森林與 XGBoost 之加密貨幣短線交易訊號預測
(Cryptocurrency Short-term Trading Signal Prediction Based on Random Forest and XGBoost)

相關論文

★ 利用資料探勘技術建立商用複合機銷售預測模型	★ 應用資料探勘技術於資源配置預測之研究-以某電腦代工支援單位為例
★ 資料探勘技術應用於航空業航班延誤分析-以C公司為例	★ 全球供應鏈下新產品的安全控管-以C公司為例
★ 資料探勘應用於半導體雷射產業-以A公司為例	★ 應用資料探勘技術於空運出口貨物存倉時間預測-以A公司為例
★ 使用資料探勘分類技術優化YouBike運補作業	★ 特徵屬性篩選對於不同資料類型之影響
★ 資料探勘應用於B2B網路型態之企業官網研究-以T公司為例	★ 衍生性金融商品之客戶投資分析與建議-整合分群與關聯法則技術
★ 應用卷積式神經網路建立肝臟超音波影像輔助判別模型	★ 基於卷積神經網路之身分識別系統
★ 能源管理系統電能補值方法誤差率比較分析	★ 企業員工情感分析與管理系統之研發
★ 資料淨化於類別不平衡問題: 機器學習觀點	★ 資料探勘技術應用於旅客自助報到之分析—以C航空公司為例

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2030-6-1以後開放)

摘要(中)

本研究旨在應用機器學習技術於加密貨幣短線交易預測，針對市場特性迥異的比特幣（Bitcoin, BTC）與狗狗幣（Dogecoin, DOGE）進行價格變動與交易訊號之分析與預測。研究整合技術指標、期貨市場微結構資料與情緒因子（恐懼與貪婪指數），建構小時級高頻資料集，並採用隨機森林（Random Forest）與極端梯度提升樹（XGBoost）兩種監督式學習模型，搭配滑動視窗與特徵排序策略，強化模型辨識短期市場變動的能力。
實驗結果顯示，XGBoost 在高波動環境下具備較佳穩定性，並且在 DOGE 模型中納入情緒變數後，可顯著提升其預測精度與 Recall 表現；相對而言，BTC 模型則主要依賴技術面與市場結構特徵，顯示其價格行為較不受情緒影響。研究成果驗證了將情緒指標納入模型可有效提升對迷因幣等情緒驅動型資產的掌握力。
本研究不僅強調模型預測準確性之提升，更著重於其實務應用價值，包含即時交易訊號產生、風險控管與量化交易策略建構之可行性。期望本研究能提供投資人與交易系統開發者具體且可行的資料驅動預測工具，促進短線操作決策之科學化與自動化。

摘要(英)

This study applies machine learning techniques to short-term cryptocur-rency trading prediction, focusing on two assets with distinct market character-istics: Bitcoin (BTC) and Dogecoin (DOGE). By integrating technical indicators, futures market microstructure data, and sentiment metrics (e.g., Fear & Greed Index), a high-frequency hourly dataset was constructed. Two supervised learning models—Random Forest and XGBoost—were implemented with a sliding window mechanism and feature ranking to enhance responsiveness to rapid market shifts.
Experimental results demonstrate that XGBoost outperforms in volatile scenarios and achieves notable improvements in precision and recall for DOGE when sentiment indicators are included. Conversely, BTC prediction perfor-mance is primarily driven by technical and structural features, indicating its rela-tive independence from emotional factors. The findings validate the effective-ness of incorporating sentiment variables for assets with strong psychological dynamics.
Beyond predictive accuracy, this study highlights the practical applicability of these models in real-world settings—supporting real-time signal generation, short-term trade execution, and risk-aware quantitative strategies. The pro-posed framework offers actionable insights and tools for investors and devel-opers aiming to systematize and automate decision-making in highly dynamic cryptocurrency markets.

關鍵字(中)

★ 機器學習
★ 短線交易
★ 隨機森林
★ 比特幣
★ 狗狗幣
★ 情緒分析

關鍵字(英)

★ Machine Learning
★ Short-Term Trading
★ XGBoost
★ Random Forest
★ Bitcoin
★ Dogecoin

論文目次

目錄
摘要 i
Abstract ii
致謝 iii
目錄 iv
表目錄 vi
圖目錄 vii
第一章緒論 1
1.1 研究背景 1
1.2 研究動機 3
1.3 研究目的 6
第二章文獻探討 7
2.1 加密貨幣市場與短線交易 7
2.2 現有研究：機器學習應用於加密貨幣市場的短線交易 8
2.3 恐懼與貪婪指數(Fear & Greed Index, FGI) 13
2.4 相關文獻回顧 14
2.5 研究缺口 17
2.6 小結 18
第三章研究方法 19
3.1 研究架構 19
3.2 資料集來源 22
3.3 資料整合與預處理 24
3.4 特徵選取 27
3.5 實驗設計 30
3.6 評估指標 31
3.7 實驗環境 33
第四章實驗結果分析 34
4.1 無滑動視窗實驗結果分析 34
4.2 有滑動視窗實驗結果分析 40
4.3 模型比較：Random Forest vs XGBoost 43
第五章結論 46
5.1 研究結果與貢獻 46
5.2 研究限制 47
5.3 未來研究方向 48
參考文獻 50

參考文獻

[1] S. Nakamoto,「Bitcoin: A Peer-to-Peer Electronic Cash System」, SSRN Electron. J., 2008, doi: 10.2139/ssrn.3440802.
[2] D. G. Baur, K. Hong及A. D. Lee,「Bitcoin: Medium of exchange or spec-ulative assets?」, J. Int. Financ. Mark. Inst. Money, 卷 54, 頁 177–189, 5月 2018, doi: 10.1016/j.intfin.2017.12.004.
[3] I. Makarov及A. Schoar,「Trading and arbitrage in cryptocurrency mar-kets」, J. Financ. Econ., 卷 135, 期 2, 頁 293–319, 2月 2020, doi: 10.1016/j.jfineco.2019.07.001.
[4] L. Kristoufek,「What Are the Main Drivers of the Bitcoin Price? Evidence from Wavelet Coherence Analysis」, PLOS ONE, 卷 10, 期 4, 頁 e0123923, 4月 2015, doi: 10.1371/journal.pone.0123923.
[5] M. Balcilar, E. Bouri, R. Gupta及D. Roubaud,「Can volume predict Bitcoin returns and volatility? A quantiles-based approach」, Econ. Model., 卷 64, 頁 74–81, 8月 2017, doi: 10.1016/j.econmod.2017.03.019.
[6] P. M. Krafft, N. D. Penna及A. Pentland,「An Experimental Study of Cryptocurrency Market Dynamics」, 收入 Proceedings of the 2018 CHI Con-ference on Human Factors in Computing Systems, 4月 2018, 頁 1–13. doi: 10.1145/3173574.3174179.
[7] T. Aste,「Cryptocurrency market structure: connecting emotions and economics」, 2019年3月3日, arXiv: arXiv:1903.00472. doi: 10.48550/arXiv.1903.00472.
[8] S. McNally, J. Roche及S. Caton, 「Predicting the Price of Bitcoin Using Machine Learning」, 收入 2018 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing(PDP), Cambridge: IEEE, 3月 2018, 頁 339–343. doi: 10.1109/PDP2018.2018.00060.
[9] Z. Jiang及J. Liang,「Cryptocurrency portfolio management with deep reinforcement learning」, 收入 2017 Intelligent Systems Conference(IntelliSys), London: IEEE, 9月 2017, 頁 905–913. doi: 10.1109/IntelliSys.2017.8324237.
[10] D. C. A. Mallqui及R. A. S. Fernandes,「Predicting the direction, maximum, minimum and closing prices of daily Bitcoin exchange rate using machine learning techniques」, Appl. Soft Comput., 卷 75, 頁 596–606, 2月 2019, doi: 10.1016/j.asoc.2018.11.038.
[11] O. Kraaijeveld及J. De Smedt,「The predictive power of public Twitter sentiment for forecasting cryptocurrency prices」, J. Int. Financ. Mark. Inst. Money, 卷 65, 頁 101188, 3月 2020, doi: 10.1016/j.intfin.2020.101188.
[12] M. Wątorek, M. Skupień, J. Kwapień及S. Drożdż,「Decomposing cryp-tocurrency high-frequency price dynamics into recurring and noisy compo-nents」, 2023, doi: 10.48550/ARXIV.2306.17095.
[13] Binance Research,「加密貨幣市場中的大規模清算：數據與影響力分析」. [線上]. 載於: https://www.binance.com/zh-TC/square/post/21381911931450
[14] Mitrade Insights, 「如何有效進行比特幣短線交易？淺談短線策略及實戰經驗分享」. [線上]. 載於: https://www.mitrade.com/cn/insights/crypto/bitcoin/trade-coin-effectively
[15] H. Sebastião及P. Godinho, 「Forecasting and trading cryptocurrencies with machine learning under changing market conditions」, Financ. Innov., 卷 7, 期 1, 頁 3, 1月 2021, doi: 10.1186/s40854-020-00217-x.
[16] A. Jabbar及S. Q. Jalil, 「A Comprehensive Analysis of Machine Learning Models for Algorithmic Trading of Bitcoin」, 2024年7月9日, arXiv: arXiv:2407.18334. doi: 10.48550/arXiv.2407.18334.
[17] S. Inder及S. Sharma, 「Predicting the Movement of Cryptocurrency “Bitcoin” Using Random Forest」, 收入 Data Science and Computational In-telligence, 卷 1483, K. R. Venugopal, P. D. Shenoy, R. Buyya, L. M. Patnaik及S. S. Iyengar, 編輯, 收入 Communications in Computer and Information Science, vol. 1483. , Cham: Springer International Publishing, 2021, 頁 166–180. doi: 10.1007/978-3-030-91244-4_14.
[18] F. Valencia, A. Gómez-Espinosa及B. Valdés-Aguirre, 「Price Movement Prediction of Cryptocurrencies Using Sentiment Analysis and Machine Learn-ing」, Entropy, 卷 21, 期 6, 頁 589, 6月 2019, doi: 10.3390/e21060589.
[19] F. Orte, J. Mira, M. J. Sánchez及P. Solana, 「A random forest-based model for crypto asset forecasts in futures markets with out-of-sample predic-tion」, Res. Int. Bus. Finance, 卷 64, 頁 101829, 1月 2023, doi: 10.1016/j.ribaf.2022.101829.
[20] S. A. Basher及P. Sadorsky, 「Forecasting Bitcoin price direction with random forests: How important are interest rates, inflation, and market volatili-ty?」, Mach. Learn. Appl., 卷 9, 頁 100355, 9月 2022, doi: 10.1016/j.mlwa.2022.100355.
[21] T. Chen及C. Guestrin, 「XGBoost: A Scalable Tree Boosting System」, 收入 Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 8月 2016, 頁 785–794. doi: 10.1145/2939672.2939785.
[22] A. Hafid, M. Ebrahim, A. Alfatemi, M. Rahouti及D. Oliveira, 「Cryptocurrency Price Forecasting Using XGBoost Regressor and Technical Indicators」, 2024年7月16日, arXiv: arXiv:2407.11786. doi: 10.48550/arXiv.2407.11786.
[23] A. Hafid, M. Rahouti, L. Kong, M. Ebrahim及M. A. Serhani, 「Predicting Bitcoin Market Trends with Enhanced Technical Indicator Integration and Classi-fication Models」, 2024年10月9日, arXiv: arXiv:2410.06935. doi: 10.48550/arXiv.2410.06935.
[24] J. Wu, X. Guo, M. Fang及J. Zhang, 「Short term return prediction of cryptocurrency based on XGBoost algorithm」, 收入 2022 International Con-ference on Big Data, Information and Computer Network(BDICN), Sanya, China: IEEE, 1月 2022, 頁 39–42. doi: 10.1109/BDICN55575.2022.00015.
[25] A. A. Oyedele, A. O. Ajayi, L. O. Oyedele, S. A. Bello及K. O. Jimoh, 「Performance evaluation of deep learning and boosted trees for cryptocur-rency closing price prediction」, Expert Syst. Appl., 卷 213, 頁 119233, 3月 2023, doi: 10.1016/j.eswa.2022.119233.
[26] J. Drahokoupil, 「Application of the XGBoost algorithm and Bayesian optimization for the Bitcoin price prediction during the COVID-19 periodJ. Drahokoupil」, FFA Work. Pap. Vol 4, 期 2022.006, 2022.
[27] Y. Zhu等, 「Price Prediction of Bitcoin Based on Adaptive Feature Selec-tion and Model Optimization」, Mathematics, 卷 11, 期 6, 頁 1335, 3月 2023, doi: 10.3390/math11061335.
[28] M. B. Osman, C. Urom, K. Guesmi及R. Benkraiem, 「Economic sentiment and the cryptocurrency market in the post-COVID-19 era」, Int. Rev. Financ. Anal., 卷 91, 頁 102962, 1月 2024, doi: 10.1016/j.irfa.2023.102962.
[29] Bitstamp, 「Understanding Market Sentiment in Crypto Trading」. [線上]. 載於: https://www.bitstamp.net/learn/crypto-trading/understanding-market-sentiment-in-crypto-trading/
[30] B. Gaies, M. S. Nakhli, J.-M. Sahut及D. Schweizer, 「Interactions between investors’ fear and greed sentiment and Bitcoin prices」, North Am. J. Econ. Finance, 卷 67, 頁 101924, 7月 2023, doi: 10.1016/j.najef.2023.101924.
[31] Coinmetro, 「Crypto Market Sentiment Indicators: Beyond the Fear and Greed Index」. [線上]. 載於: https://www.coinmetro.com/learning-lab/crypto-market-sentiment-indicators
[32] J.-N. Wang, H.-C. Liu及Y.-T. Hsu, 「A U-shaped relationship between the crypto fear-greed index and the price synchronicity of cryptocurrencies」, Fi-nance Res. Lett., 卷 59, 頁 104763, 1月 2024, doi: 10.1016/j.frl.2023.104763.
[33] Y. Huang, T. Xu, C. Xue及J. Zhang, 「How does the Bitcoin Sentiment Index of Fear & Greed affect Bitcoin returns?」, Corp. Ownersh. Control, 卷 21, 期 2, 頁 121–131, 2024, doi: 10.22495/cocv21i2art10.
[34] M. Mudassir, S. Bennbaia, D. Unal及M. Hammoudeh, 「Time-series forecasting of Bitcoin prices using high-dimensional features: a machine learn-ing approach」, Neural Comput. Appl., 7月 2020, doi: 10.1007/s00521-020-05129-6.
[35] J.-Z. Huang, W. Huang及J. Ni, 「Predicting bitcoin returns using high-dimensional technical indicators」, J. Finance Data Sci., 卷 5, 期 3, 頁 140–155, 9月 2019, doi: 10.1016/j.jfds.2018.10.001.
[36] L. Alessandretti, A. ElBahrawy, L. M. Aiello及A. Baronchelli, 「Anticipating cryptocurrency prices using machine learning」, 2018, doi: 10.48550/ARXIV.1805.08550.
[37] R. C. Phillips及D. Gorse, 「Cryptocurrency price drivers: Wavelet co-herence analysis revisited」, PLOS ONE, 卷 13, 期 4, 頁 e0195200, 4月 2018, doi: 10.1371/journal.pone.0195200.
[38] I. E. Livieris, E. Pintelas, S. Stavroyiannis, and P. Pintelas, “Ensemble Deep Learning Models for Forecasting Cryptocurrency Time-Series,” Algorithms, vol. 13, no. 5, p. 121, May 2020, doi: https://doi.org/10.3390/a13050121.
[39] V. Saha, “Predicting Future Cryptocurrency Prices Using Machine Learning Algorithms,” Journal of Data Analysis and Information Processing, vol. 11, no. 4, pp. 400–419, Sep. 2023, doi: https://doi.org/10.4236/jdaip.2023.114021.
[40] H. M. Tanrikulu and H. Pabuccu, “The Effect of Data Types’ on the Performance of Machine Learning Algorithms for Cryptocurrency Predic-tion,” Computational Economics, Mar. 2025, doi: https://doi.org/10.1007/s10614-025-10919-y.
[41] A. S. Alshehri, “Predicting Cryptocurrency Returns Using Classification and Regression Machine Learning Model,” Journal of Electrical Systems, vol. 20, no. 4s, pp. 539–553, Apr. 2024, doi: https://doi.org/10.52783/jes.2065.

指導教授

蔡志豐(Chih-Fong Tsai)

審核日期

2025-6-20

推文