簡易檢索 / 詳目顯示

回結果列表
研究生： 陳佳詳
Chen, Chia-Hxiang
論文名稱： 應用於日本期貨市場基於 FPGA 高頻交易硬體策略
FPGA-Based High-Frequency Trading Hardware Strategy for the Japanese Futures Market
指導教授： 馬席彬
Ma, Hsi-Pin
口試委員: 黃稚存
Huang, Chih-Tsun
蔡佩芸
Tsai, Pei-Yun
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 電機工程學系
Department of Electrical Engineering
論文出版年： 2025
畢業學年度： 113
語文別： 中文
論文頁數： 82
中文關鍵詞： 高頻交易 、交易策略 、硬體加速
外文關鍵詞： High Frequency Trading, Trading strategies, Hardware acceleration
相關次數： 點閱：46  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 現代金融市場中，交易者廣泛應用各種演算法來進行交易決策，並基於歷史交易數據嘗試預測未來的市場走向。然而，高頻交易的興起顯著增加了市場的短期波動性與不可預測性。而儘管已有研究探索基於場效可程式化邏輯閘陣列（FPGA）的高頻交易系統，但以硬體實現的高頻交易策略卻相較少見。

    本論文提出一種基於FPGA實現的硬體高頻交易策略，結合FPGA低延遲的優勢與逐筆訂單簿的詳細行情資訊，有效避免因高速下單特性造成的市場信號欺騙性問題。此策略透過誘餌限價單緊跟市場動向，在大趨勢瞬間於對應月份開倉組建跨月套利部位，並利用交易所提供的跨月價差商品完成平倉。藉由低風險、高成功率的套利模式，策略旨在穩定剝取利潤。此外，基於模擬市場環境和系統延遲的綜合分析，策略進一步設計了掛單量閥值以降低交易風險。

    最終回測結果顯示，該跨月價差策略取得了理想表現。誘餌放置於近月的策略達到 65.22% 的勝率，而誘餌放置於遠月的策略則取得了 78.76% 的勝率，展現了策略在實際應用中的穩定性與有效性。

    In modern financial markets, traders increasingly rely on algorithms to guide trading decisions and predict market trends using historical transaction data. However, the growth of high-frequency trading (HFT) has amplified short-term market volatility and unpredictability. Although prior studies have explored FPGA-based HFT systems, strategies directly implemented in hardware remain relatively rare.

    This paper presents an FPGA-based HFT strategy leveraging FPGA's low-latency advantages and detailed tick-by-tick order book data. The approach counters deceptive signals caused by high-speed order placement from competing traders. By deploying bait limit orders to align with market trends, the strategy identifies significant movements and establishes cross-month arbitrage positions. The closing of positions is executed using exchange-provided calendar spread products. Designed as a low-risk, high-success-rate model, the strategy aims for consistent profitability. Simulated market environments and latency analyses further refine order quantity thresholds to manage risks effectively.

    Backtesting results demonstrate the strategy's reliability. The bait placement strategy for near-month contracts achieved a 65.22% success rate, while far-month contracts achieved 78.76%, validating its stability and real-world applicability.

    誌謝 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 摘要 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III 第 一 章 緒論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 研究背景 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 研究動機 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 論文大綱 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 第 二 章 文獻回顧 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1 金融行情數據 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.1 限價單與市價單 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.2 限價訂單簿 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2 金融技術分析 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.3 高頻交易系統 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.4 文獻回顧與分析 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 第 三 章 研究方法 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.1 跨月價差策略 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.1.1 標準組合商品 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.1.2 隱含訂單 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.1.3 套利空間 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.1.4 策略邏輯 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.1.5 誘餌位置與回打操作方案 . . . . . . . . . . . . . . . . . . . . . . . 22 3.1.6 回打 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.1.7 誘餌 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.1.8 衍伸操作 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.2 平倉流程 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.2.1 趨勢指標 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.2.2 平倉策略 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.3 延遲對策略的影響 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.4 硬體架構設計 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.4.1 誘餌回打模組 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.4.2 平倉模組 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.4.3 趨勢指標計算器 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 第 四 章 實驗結果與討論 . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.1 實驗環境 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.2 硬體模擬 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.2.1 歷史數據 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.2.2 模擬交易所模組 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 4.2.3 模擬流程 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 V VI 目錄 4.3 硬體使用率 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.4 獲利回測及分析 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.4.1 測試環境 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.4.2 評估指標 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.4.3 策略比較 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.4.4 實驗結果 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 第 五 章 結論與未來規劃 . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.1 結論 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.2 未來規劃 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 參考文獻 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

    [1] I. Aldridge,
    High-frequency trading: a practical guide to algorithmic strategies and trading
    systems. John Wiley & Sons, 2013, vol. 604.
    [2] C. Leber, B. Geib, and H. Litz, “High frequency trading acceleration using fpgas,” in
    2011
    21st International Conference on Field Programmable Logic and Applications, Sep. 2011,
    pp. 317–322.
    [3] F. E. Tay and L. Cao, “Application of support vector machines in financial time series
    forecasting,”
    omega, vol. 29, no. 4, pp. 309–317, 2001.
    [4] B. M. Henrique, V. A. Sobreiro, and H. Kimura, “Literature review: Machine learning
    techniques applied to financial market prediction,”
    Expert Systems with Applications,
    vol. 124, pp. 226–251, 2019. [Online]. Available: https://www.sciencedirect.com/
    science/article/pii/S095741741930017X
    [5] R. Cooper, M. Davis, and B. V. Vliet, “The mysterious ethics of high-frequency trading,”Business Ethics Quarterly, vol. 26, no. 1, pp. 1–22, 2016.
    [6] C. M. Jones, “What do we know about high-frequency trading?” 2013, columbia Business
    School Research Paper No. 13-11. [Online]. Available: https://ssrn.com/abstract=2236201
    [7] A. Ye, V. L. R. Chinthalapati, A. Serguieva, and E. Tsang, “Developing sustainable trading
    strategies using directional changes with high frequency data,” in
    2017 IEEE International
    Conference on Big Data (Big Data), 2017, pp. 4265–4271.
    [8] C.-I. Lee, “The price discovery of futures spread,” https://etd.lib.nycu.edu.tw/cgi-bin/
    gs32/hugsweb.cgi?o=dnthucdr&s=id=%22GH029971513%22.&searchmode=basic,
    2012.
    [9] L. E. Harris and V. Panchapagesan, “The information content of the limit order book:
    evidence from nyse specialist trading decisions,”
    Journal of Financial Markets, vol. 8,
    no. 1, pp. 25–67, 2005. [Online]. Available: https://www.sciencedirect.com/science/
    article/pii/S1386418104000242
    [10] W. Cheng, S.-c. Liu, H.-y. Jiao, and W.-h. Qiu, “How does limit order book information
    affect trading strategy and market quality: Simulations of an agent-based stock market,”
    in
    2009 International Conference on Management and Service Science, 2009, pp. 1–4.
    [11] I. Svetunkov and F. Petropoulos, “Old dog, new tricks: a modelling view of simple moving
    averages,”
    International Journal of Production Research, vol. 56, no. 18, pp. 6034–6047,
    2018.
    [12] S. Hansun, “A new approach of moving average method in time series analysis,” in
    2013
    Conference on New Media Studies (CoNMedia), 2013, pp. 1–4.
    [13] D. S. Grebenkov and J. Serror, “Following a trend with an exponential moving
    average: Analytical results for a gaussian model,”
    Physica A: Statistical Mechanics
    and its Applications, vol. 394, pp. 288–303, 2014. [Online]. Available: https:
    //www.sciencedirect.com/science/article/pii/S0378437113009722
    [14] A. Ţăran-Moroşan, “The relative strength index revisited,”
    African Journal of Business
    Management, vol. 5, no. 14, pp. 5855–5862, 2011.
    [15] W. W. H. Tsang, T. T. L. Chong
    et al., “Profitability of the on-balance volume indicator,”Economics Bulletin, vol. 29, no. 3, pp. 2424–2431, 2009.
    [16] M. El-Shahawi, A. Hamza, A. Bashammakh, and W. Al-Saggaf, “An overview
    on the accumulation, distribution, transformations, toxicity and analytical methods
    for the monitoring of persistent organic pollutants,”
    Talanta, vol. 80, no. 5, pp.
    1587–1597, 2010. [Online]. Available: https://www.sciencedirect.com/science/article/
    pii/S0039914009007620
    [17] D.-H. Zhao, “A hardware-accelerated modular algorithmic trading sys-
    tem,” https://etd.lib.nycu.edu.tw/cgi-bin/gs32/hugsweb.cgi?o=dnthucdr&s=id=
    %22G021090616030%22.&searchmode=basic, 2022.
    [18] Y.-C. Kao, H.-A. Chen, and H.-P. Ma, “An fpga-based high-frequency trading system for
    10 gigabit ethernet with a latency of 433 ns,” in
    2022 International Symposium on VLSI
    Design, Automation and Test (VLSI-DAT), 2022, pp. 1–4.
    [19] A. Kablan, “Adaptive neuro fuzzy inference systems for high frequency financial trad-
    ing and forecasting,” in
    2009 Third International Conference on Advanced Engineering
    Computing and Applications in Sciences, 2009, pp. 105–110.
    [20] Y.-T. Liu, “Trading agents based on deep reinforcement learning with tick
    data,” https://etd.lib.nycu.edu.tw/cgi-bin/gs32/hugsweb.cgi?o=dnthucdr&s=id=
    %22G021090616390%22.&searchmode=basic, 2022.
    [21] A. Kablan and W. Ng, “High frequency trading strategy using the hilbert transform,” inThe 6th International Conference on Networked Computing and Advanced Information
    Management, 2010, pp. 466–471.
    [22] B. Dai, B. R. Marshall, N. H. Nguyen, and N. Visaltanachoti, “Risk reduction using
    trailing stop-loss rules,”
    International Review of Finance, vol. 21, no. 4, pp. 1334–1352,
    2021. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1111/irfi.12328
    [23]
    J-GATE3.0 OUCH Connectivity Manual V1.6, Japan Exchange Group (JPX), 2016.
    [Online]. Available: https://www.jpx.co.jp/english
    [24]
    J-GATE3.0 ITCH Connectivity Manual V1.8, Japan Exchange Group (JPX), 2018.
    [Online]. Available: https://www.jpx.co.jp/english
    [25] “Genium INET® Nasdaq transport protocol overview core,” NASDAQ, 2020.

    QR CODE