1. NCCU Academic Hub
  2. Publications
  3. College of Commerce
  4. Theses
  5. Item

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 基於限價訂單簿資料之整點前後市場流動性變化分析與預測－以比特幣為例
Analyzing and Forecasting Hourly Liquidity Shifts Using Limit Order Book Data: Evidence from Bitcoin
作者 黃柏翔
Huang, Po-Hsiang
貢獻者 羅秉政
Kendro Vincent
黃柏翔
Huang, Po-Hsiang
關鍵詞 限價訂單簿
流動性預測
機器學習
LOB
Liquidity prediction
Machine learning
日期 2025
上傳時間 4-Aug-2025 14:32:21 (UTC+8)
摘要 　　本研究以比特幣限價訂單簿資料為基礎，探討每小時整點前後市場微結構中流動性指標：買賣價差（Bid-Ask Spread）與訂單失衡程度（Order imbalance）之變化與可預測性。透過自行錄製之高頻訂單簿資料，進行統計檢定證實整點前後指標存在顯著差異，顯示整點為潛在市場結構轉折點。預測任務方面，分別針對整點後五秒內之平均價差、平均訂單失衡程度與失衡方向建構迴歸與分類模型，結果顯示 LSTM 模型在分類任務中表現最佳（F1-score = 0.56），並於迴歸任務中取得最低 RMSE。進一步以隨機森林分析特徵重要性，顯示短期訂單失衡程度與深度相關指標對預測最具貢獻。研究結果顯示，加密貨幣市場雖無明確開收盤制度，整點時點仍反映出顯著的流動性變化，並可用以輔助短期交易策略設計。
This study investigates the changes and predictability of market microstructure around hourly intervals in the Bitcoin market, focusing on limit order book indicators such as bid-ask spread and order imbalance. Using self-collected high-frequency order book data from Binance, statistical tests confirm significant differences in liquidity measures before and after each hourly close, suggesting structural shifts around the clock. Prediction tasks are designed to forecast the average spread, average order imbalance, and imbalance direction in the five seconds after each hourly mark. Results show that the LSTM model achieves the best performance in the classification task (F1-score = 0.56) and lowest RMSE in regression tasks. Feature importance analysis using random forests highlights the predictive value of short-term imbalance and depth-related features. These findings indicate that despite the absence of formal closing times in cryptocurrency markets, strategic microstructure shifts occur at hourly intervals and can inform short-term trading strategy design.
參考文獻 Brauneis, A., Mestel, R., & Theissen, E. (2025). The crypto world trades at tea time: intraday evidence from centralized exchanges across the globe. Review of Quantitative Finance and Accounting, 64(1), 275-304. Brock, W. A., & Kleidon, A. W. (1992). Periodic market closure and trading volume: A model of intraday bids and asks. Journal of Economic Dynamics and Control, 16(3-4), 451-489. Chordia, T., Roll, R., & Subrahmanyam, A. (2002). Order imbalance, liquidity, and market returns. Journal of Financial economics, 65(1), 111-130. Cont, R., Kukanov, A., & Stoikov, S. (2014). The price impact of order book events. Journal of financial econometrics, 12(1), 47-88. Easley, D., Hvidkjaer, S., & O'hara, M. (2002). Is information risk a determinant of asset returns?. The journal of finance, 57(5), 2185-2221. Glosten, L. R., & Milgrom, P. R. (1985). Bid, ask and transaction prices in a specialist market with heterogeneously informed traders. Journal of financial economics, 14(1), 71-100. Hansen, P. R., Kim, C., & Kimbrough, W. (2024). Periodicity in cryptocurrency volatility and liquidity. Journal of Financial Econometrics, 22(1), 224-251. Heston, S. L., Korajczyk, R. A., & Sadka, R. (2010). Intraday patterns in the cross‐section of stock returns. The Journal of Finance, 65(4), 1369-1407. Mazza, P. (2015). Price dynamics and market liquidity: An intraday event study on Euronext. The Quarterly Review of Economics and Finance, 56, 139-153. Muravyev, D., & Picard, J. (2022). Does trade clustering reduce trading costs? Evidence from periodicity in algorithmic trading. Financial Management, 51(4), 1201-1229. Nousi, P., Tsantekidis, A., Passalis, N., Ntakaris, A., Kanniainen, J., Tefas, A., ... & Iosifidis, A. (2019). Machine learning for forecasting mid-price movements using limit order book data. Ieee Access, 7, 64722-64736. Roşu, I. (2009). A dynamic model of the limit order book. The Review of Financial Studies, 22(11), 4601-4641. Sirignano, J., & Cont, R. (2021). Universal features of price formation in financial markets: perspectives from deep learning. In Machine learning and AI in finance (pp. 5-15). Routledge. Wang, J. N., Liu, H. C., & Hsu, Y. T. (2020). Time-of-day periodicities of trading volume and volatility in Bitcoin exchange: does the stock market matter?. Finance Research Letters, 34, 101243. Wątorek, M., Skupień, M., Kwapień, J., & Drożdż, S. (2023). Decomposing cryptocurrency high-frequency price dynamics into recurring and noisy components. Chaos: An Interdisciplinary Journal of Nonlinear Science, 33(8).
描述 碩士
國立政治大學
金融學系
112352012
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0112352012
資料類型 thesis
dc.contributor.advisor 羅秉政zh_TW
dc.contributor.advisor Kendro Vincenten_US
dc.contributor.author (Authors) 黃柏翔zh_TW
dc.contributor.author (Authors) Huang, Po-Hsiangen_US
dc.creator (作者) 黃柏翔zh_TW
dc.creator (作者) Huang, Po-Hsiangen_US
dc.date (日期) 2025en_US
dc.date.accessioned 4-Aug-2025 14:32:21 (UTC+8)-
dc.date.available 4-Aug-2025 14:32:21 (UTC+8)-
dc.date.issued (上傳時間) 4-Aug-2025 14:32:21 (UTC+8)-
dc.identifier (Other Identifiers) G0112352012en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/158589-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 金融學系zh_TW
dc.description (描述) 112352012zh_TW
dc.description.abstract (摘要) 　　本研究以比特幣限價訂單簿資料為基礎，探討每小時整點前後市場微結構中流動性指標：買賣價差（Bid-Ask Spread）與訂單失衡程度（Order imbalance）之變化與可預測性。透過自行錄製之高頻訂單簿資料，進行統計檢定證實整點前後指標存在顯著差異，顯示整點為潛在市場結構轉折點。預測任務方面，分別針對整點後五秒內之平均價差、平均訂單失衡程度與失衡方向建構迴歸與分類模型，結果顯示 LSTM 模型在分類任務中表現最佳（F1-score = 0.56），並於迴歸任務中取得最低 RMSE。進一步以隨機森林分析特徵重要性，顯示短期訂單失衡程度與深度相關指標對預測最具貢獻。研究結果顯示，加密貨幣市場雖無明確開收盤制度，整點時點仍反映出顯著的流動性變化，並可用以輔助短期交易策略設計。zh_TW
dc.description.abstract (摘要) This study investigates the changes and predictability of market microstructure around hourly intervals in the Bitcoin market, focusing on limit order book indicators such as bid-ask spread and order imbalance. Using self-collected high-frequency order book data from Binance, statistical tests confirm significant differences in liquidity measures before and after each hourly close, suggesting structural shifts around the clock. Prediction tasks are designed to forecast the average spread, average order imbalance, and imbalance direction in the five seconds after each hourly mark. Results show that the LSTM model achieves the best performance in the classification task (F1-score = 0.56) and lowest RMSE in regression tasks. Feature importance analysis using random forests highlights the predictive value of short-term imbalance and depth-related features. These findings indicate that despite the absence of formal closing times in cryptocurrency markets, strategic microstructure shifts occur at hourly intervals and can inform short-term trading strategy design.en_US
dc.description.tableofcontents 第一章　緒論 1 第二章　文獻回顧 3 　第一節　流動性與價格 3 　第二節　傳統金融市場的時間結構現象 4 　第三節　加密貨幣市場的時間結構現象 5 第三章　研究方法 7 　第一節　資料來源與格式 8 　第二節　流動性指標與統計檢定方法 10 　第三節　流動性預測 13 　第四節　評估函數 24 第四章　實證結果與討論 26 　第一節　整點前後的流動性變化：統計檢定結果 26 　第二節　迴歸任務分析：預測平均買賣價差 27 第三節　迴歸任務分析：預測平均買賣價差 29 　第四節　分類任務分析：預測訂單簿失衡方向 30 　第五節　特徵重要性分析 32 第五章　結論與建議 35 　第一節　研究結論 36 　第二節　研究貢獻 37 　第三節　研究限制與未來方向 38 參考文獻 40zh_TW
dc.format.extent 1603071 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0112352012en_US
dc.subject (關鍵詞) 限價訂單簿zh_TW
dc.subject (關鍵詞) 流動性預測zh_TW
dc.subject (關鍵詞) 機器學習zh_TW
dc.subject (關鍵詞) LOBen_US
dc.subject (關鍵詞) Liquidity predictionen_US
dc.subject (關鍵詞) Machine learningen_US
dc.title (題名) 基於限價訂單簿資料之整點前後市場流動性變化分析與預測－以比特幣為例zh_TW
dc.title (題名) Analyzing and Forecasting Hourly Liquidity Shifts Using Limit Order Book Data: Evidence from Bitcoinen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) Brauneis, A., Mestel, R., & Theissen, E. (2025). The crypto world trades at tea time: intraday evidence from centralized exchanges across the globe. Review of Quantitative Finance and Accounting, 64(1), 275-304. Brock, W. A., & Kleidon, A. W. (1992). Periodic market closure and trading volume: A model of intraday bids and asks. Journal of Economic Dynamics and Control, 16(3-4), 451-489. Chordia, T., Roll, R., & Subrahmanyam, A. (2002). Order imbalance, liquidity, and market returns. Journal of Financial economics, 65(1), 111-130. Cont, R., Kukanov, A., & Stoikov, S. (2014). The price impact of order book events. Journal of financial econometrics, 12(1), 47-88. Easley, D., Hvidkjaer, S., & O'hara, M. (2002). Is information risk a determinant of asset returns?. The journal of finance, 57(5), 2185-2221. Glosten, L. R., & Milgrom, P. R. (1985). Bid, ask and transaction prices in a specialist market with heterogeneously informed traders. Journal of financial economics, 14(1), 71-100. Hansen, P. R., Kim, C., & Kimbrough, W. (2024). Periodicity in cryptocurrency volatility and liquidity. Journal of Financial Econometrics, 22(1), 224-251. Heston, S. L., Korajczyk, R. A., & Sadka, R. (2010). Intraday patterns in the cross‐section of stock returns. The Journal of Finance, 65(4), 1369-1407. Mazza, P. (2015). Price dynamics and market liquidity: An intraday event study on Euronext. The Quarterly Review of Economics and Finance, 56, 139-153. Muravyev, D., & Picard, J. (2022). Does trade clustering reduce trading costs? Evidence from periodicity in algorithmic trading. Financial Management, 51(4), 1201-1229. Nousi, P., Tsantekidis, A., Passalis, N., Ntakaris, A., Kanniainen, J., Tefas, A., ... & Iosifidis, A. (2019). Machine learning for forecasting mid-price movements using limit order book data. Ieee Access, 7, 64722-64736. Roşu, I. (2009). A dynamic model of the limit order book. The Review of Financial Studies, 22(11), 4601-4641. Sirignano, J., & Cont, R. (2021). Universal features of price formation in financial markets: perspectives from deep learning. In Machine learning and AI in finance (pp. 5-15). Routledge. Wang, J. N., Liu, H. C., & Hsu, Y. T. (2020). Time-of-day periodicities of trading volume and volatility in Bitcoin exchange: does the stock market matter?. Finance Research Letters, 34, 101243. Wątorek, M., Skupień, M., Kwapień, J., & Drożdż, S. (2023). Decomposing cryptocurrency high-frequency price dynamics into recurring and noisy components. Chaos: An Interdisciplinary Journal of Nonlinear Science, 33(8).zh_TW