Publications-Theses
Article View/Open
Publication Export
-
題名 卷積神經網路結合技術指標交易策略在台灣加權指數期貨之應用
Applications of Trading Strategies for Convolution Neural Network and Technical Indicators in Taiwan Stock Price Index Futures作者 李杰穎
Lee, Chieh-Ying貢獻者 林士貴<br>王釧茹
Lin, Shih-Kuei<br>Wang, Chuan-Ju
李杰穎
Lee, Chieh-Ying關鍵詞 卷積神經網路
技術指標
隨機指標
布林通道
Convolutional neural networks
Technical indicator
Stochastic oscillator
Boolinger band日期 2019 上傳時間 1-Jul-2019 10:48:28 (UTC+8) 摘要 在金融科技(Financial Technology, FinTech)迅速發展之下，許多金融服務都嘗試結合最新的突破技術，例如:行動支付(轉帳)、數位銀行崛起……等等，讓我們隨時都能使用金融服務，而不需要親自再跑一趟實體銀行，由此可見，金融發展與科技息息相關，目前金融也正在朝全自動化、智能化為目標努力。在金融交易方面，已經可以達到透過技術指標建構模型自動下單的技術，本論文將卷積神經網路模型(Convolutional Neural Network, CNN)應用在技術指標交易策略上，將開盤價、最高價、收盤價、最低價及技術指標走勢轉換成圖像，希望憑藉卷積神經網路模型優異的圖像辨識能力，對獲利和虧損的交易策略進行特徵提取，達到提高交易策略準確率的目的。實證結果發現，不管交易策略在多單或是空單方面，在經過卷積神經網路模型訓練之後，都能有效的提高準確率。臺灣近期也越來越重視人工智能結合金融服務的發展，人工智能不但能達到全自動化的服務，也能帶給使用金融服務的客戶煥然一新的體驗，並達到更快速的產品和服務交付，提高金融服務的效率，本文期望透過將類神經網路結合技術指標交易策略的方法，使未來學界朝人工智能交易為目標發展之時，能夠有更多不同的想法。
In terms of financial trading, it is possible to achieve automatically placing orders through following several technical indicators. In this paper, we apply Convolutional Neural Networks to the technical trading strategy. We converted opening price, highest price, closing price, lowest price, and the trend of technical indicator into images. We hoped that by the excellent ability of image recognition of Convolutional Neural Networks, we can extract the features of profit strategies and loss strategies, and then improve the accuracy of trading strategies. The empirical results show that no matter long strategies or short strategies were used, after the training of Convolutional Neural Networks, the accuracy of gaining profits from the strategies can be improved effectively. In recent years, people in Taiwan pay more and more attention to the development of artificial intelligence combined with financial services. My expectation is to provide different ideas to scholars and experts who worked hard in this related area in this thesis, so that they might further develop new techniques in this area from different perspectives.參考文獻 [1] Akita, R., Yoshihara, A., Matsubara, T., and Uehara K., (2016). Deep Learning for Stock Prediction Using Numerical and Textual Information. Kobe University, Master’s Thesis.[2] Alexander, S. S., (1961). Price Movement in Speculative Markets: Trends or Random Walks.[3] Choudhry, R., and Garg, K., (2008). A Hybrid Machine Learning System for Stock Market Forecasting. World Academy of Science, Engineering and Technology International Journal of Computer and Information Engineering Vol:2, No:3.[4] Hsu, P.H., and Taylor, M.P., (2016). Technical Trading: Is It Still Beating The Foreign Exchange Market? Journal of International Economics, 102, 188-208.[5] Kim, K.J., (2000). Genetic Algorithms Approach to Feature Discretization in Artificial Neural Networks for The Prediction of Stock Price Index. Expert Systems with Applications, 19(2), 125–132.[6] Krizhevsky, A., Sutskever, I., and Hinton, G.E., (2012). ImageNet Classification with Deep Convolutional Neural Networks. Proceeding NIPS`12 Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1, 1097-1105.[7] LeCun, Y., and Bengio, Y., (1995). Convolutional Networks for Images, Speech, and Time Series. The Handbook of Brain Theory and Neural Networks, 255-258.[8] Levich, R., and Thomas, L., (1993). The Significance of Technical Trading Rule Profits in the Foreign Exchange Market: ABootstrap Approach. Journal of International Money and Finance, 12, 451–474.[9] Lukac, L.P., Brorsen, B.W., and Irwin, S.H., (2006). A Test of Futures Market Disequilibrium Using Twelve Different Technical Trading Systems. 623-639.[10] Persio, L.D., and Honchar, O., (2016). Artificial Neural Networks Architectures for Stock Price Prediction: Comparisons and Applications. University of Verona, Master’s Thesis.[11] Pruitt, S.W., and White, R.E., (1988). The CRISMA Trading System: Who Says Technical Analysis Can’t Beat The Market? Journal of Portfolio Management, 55-58.[12] Shen, S., Jiang, H. and Zhang, T., (2012). Stock Market Forecasting Using Machine Learning Algorithms. Stanford University, Master’s Thesis.[13] Taylor, M. P., and Allen, H.L., (1992). The Use of Technical Analysis in the Foreign Exchange Market. Journal of International Money and Finance, 11, 304-314.[14] Taylor, S. T., (1994). Trading Futures Using A Channel Rule: A Study of The Predictive Power of Technical Analysis with Currency Examples. Journal of Futures Markets, Volume 14, Issue 2.[15] Tsantekidis, A., Passalis, N., Tefas, A., Kanniainen, J., Gabbouj, M., & Iosifidis, A. (2017). Forecasting stock prices from the limit order book using convolutional neural networks. In 2017 IEEE 19th Conference on Business Informatics (CBI) (Vol. 1, pp. 7-12). IEEE.[16] 吳永樂， (2014)。運用支持向量機和決策樹預測台指期走勢，國立政治大學，碩士論文。[17] 賴嘉蔚，(2017)。卷積神經網路預測時間序列能力分析，國立政治大學，碩士論文。[18] 劉昭雨，(2017)。卷積神經網路在金融技術指標之應用，國立東華大學，碩士論文。 描述 碩士
國立政治大學
金融學系
106352034資料來源 http://thesis.lib.nccu.edu.tw/record/#G0106352034 資料類型 thesis dc.contributor.advisor 林士貴<br>王釧茹 zh_TW dc.contributor.advisor Lin, Shih-Kuei<br>Wang, Chuan-Ju en_US dc.contributor.author (Authors) 李杰穎 zh_TW dc.contributor.author (Authors) Lee, Chieh-Ying en_US dc.creator (作者) 李杰穎 zh_TW dc.creator (作者) Lee, Chieh-Ying en_US dc.date (日期) 2019 en_US dc.date.accessioned 1-Jul-2019 10:48:28 (UTC+8) - dc.date.available 1-Jul-2019 10:48:28 (UTC+8) - dc.date.issued (上傳時間) 1-Jul-2019 10:48:28 (UTC+8) - dc.identifier (Other Identifiers) G0106352034 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/124145 - 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) 106352034 zh_TW dc.description.abstract (摘要) 在金融科技(Financial Technology, FinTech)迅速發展之下，許多金融服務都嘗試結合最新的突破技術，例如:行動支付(轉帳)、數位銀行崛起……等等，讓我們隨時都能使用金融服務，而不需要親自再跑一趟實體銀行，由此可見，金融發展與科技息息相關，目前金融也正在朝全自動化、智能化為目標努力。在金融交易方面，已經可以達到透過技術指標建構模型自動下單的技術，本論文將卷積神經網路模型(Convolutional Neural Network, CNN)應用在技術指標交易策略上，將開盤價、最高價、收盤價、最低價及技術指標走勢轉換成圖像，希望憑藉卷積神經網路模型優異的圖像辨識能力，對獲利和虧損的交易策略進行特徵提取，達到提高交易策略準確率的目的。實證結果發現，不管交易策略在多單或是空單方面，在經過卷積神經網路模型訓練之後，都能有效的提高準確率。臺灣近期也越來越重視人工智能結合金融服務的發展，人工智能不但能達到全自動化的服務，也能帶給使用金融服務的客戶煥然一新的體驗，並達到更快速的產品和服務交付，提高金融服務的效率，本文期望透過將類神經網路結合技術指標交易策略的方法，使未來學界朝人工智能交易為目標發展之時，能夠有更多不同的想法。 zh_TW dc.description.abstract (摘要) In terms of financial trading, it is possible to achieve automatically placing orders through following several technical indicators. In this paper, we apply Convolutional Neural Networks to the technical trading strategy. We converted opening price, highest price, closing price, lowest price, and the trend of technical indicator into images. We hoped that by the excellent ability of image recognition of Convolutional Neural Networks, we can extract the features of profit strategies and loss strategies, and then improve the accuracy of trading strategies. The empirical results show that no matter long strategies or short strategies were used, after the training of Convolutional Neural Networks, the accuracy of gaining profits from the strategies can be improved effectively. In recent years, people in Taiwan pay more and more attention to the development of artificial intelligence combined with financial services. My expectation is to provide different ideas to scholars and experts who worked hard in this related area in this thesis, so that they might further develop new techniques in this area from different perspectives. en_US dc.description.tableofcontents 第一章 緒論 1第一節 研究動機 1第二節 研究目的 2第三節 研究背景 3第二章 文獻回顧 4第一節 技術指標交易金融市場商品之相關文獻 4第二節 機器學習預測市場商品漲跌之相關文獻 5第三節 文獻回顧總結 6第三章 研究方法 7第一節 技術分析探討 7第二節 技術指標選擇 8第三節 卷積神經網路模型(CNN)簡介 13第四節 卷積神經網路模型(CNN)發展歷程 13第五節 卷積神經網路模型(CNN)架構 15第四章 實證分析 28第一節 研究對象 28第二節 實驗架構 32第三節 實證結果 38第五章 結論與建議 52第一節 結論 52第二節 未來展望 53參考文獻 54 zh_TW dc.format.extent 2982711 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0106352034 en_US dc.subject (關鍵詞) 卷積神經網路 zh_TW dc.subject (關鍵詞) 技術指標 zh_TW dc.subject (關鍵詞) 隨機指標 zh_TW dc.subject (關鍵詞) 布林通道 zh_TW dc.subject (關鍵詞) Convolutional neural networks en_US dc.subject (關鍵詞) Technical indicator en_US dc.subject (關鍵詞) Stochastic oscillator en_US dc.subject (關鍵詞) Boolinger band en_US dc.title (題名) 卷積神經網路結合技術指標交易策略在台灣加權指數期貨之應用 zh_TW dc.title (題名) Applications of Trading Strategies for Convolution Neural Network and Technical Indicators in Taiwan Stock Price Index Futures en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) [1] Akita, R., Yoshihara, A., Matsubara, T., and Uehara K., (2016). Deep Learning for Stock Prediction Using Numerical and Textual Information. Kobe University, Master’s Thesis.[2] Alexander, S. S., (1961). Price Movement in Speculative Markets: Trends or Random Walks.[3] Choudhry, R., and Garg, K., (2008). A Hybrid Machine Learning System for Stock Market Forecasting. World Academy of Science, Engineering and Technology International Journal of Computer and Information Engineering Vol:2, No:3.[4] Hsu, P.H., and Taylor, M.P., (2016). Technical Trading: Is It Still Beating The Foreign Exchange Market? Journal of International Economics, 102, 188-208.[5] Kim, K.J., (2000). Genetic Algorithms Approach to Feature Discretization in Artificial Neural Networks for The Prediction of Stock Price Index. Expert Systems with Applications, 19(2), 125–132.[6] Krizhevsky, A., Sutskever, I., and Hinton, G.E., (2012). ImageNet Classification with Deep Convolutional Neural Networks. Proceeding NIPS`12 Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1, 1097-1105.[7] LeCun, Y., and Bengio, Y., (1995). Convolutional Networks for Images, Speech, and Time Series. The Handbook of Brain Theory and Neural Networks, 255-258.[8] Levich, R., and Thomas, L., (1993). The Significance of Technical Trading Rule Profits in the Foreign Exchange Market: ABootstrap Approach. Journal of International Money and Finance, 12, 451–474.[9] Lukac, L.P., Brorsen, B.W., and Irwin, S.H., (2006). A Test of Futures Market Disequilibrium Using Twelve Different Technical Trading Systems. 623-639.[10] Persio, L.D., and Honchar, O., (2016). Artificial Neural Networks Architectures for Stock Price Prediction: Comparisons and Applications. University of Verona, Master’s Thesis.[11] Pruitt, S.W., and White, R.E., (1988). The CRISMA Trading System: Who Says Technical Analysis Can’t Beat The Market? Journal of Portfolio Management, 55-58.[12] Shen, S., Jiang, H. and Zhang, T., (2012). Stock Market Forecasting Using Machine Learning Algorithms. Stanford University, Master’s Thesis.[13] Taylor, M. P., and Allen, H.L., (1992). The Use of Technical Analysis in the Foreign Exchange Market. Journal of International Money and Finance, 11, 304-314.[14] Taylor, S. T., (1994). Trading Futures Using A Channel Rule: A Study of The Predictive Power of Technical Analysis with Currency Examples. Journal of Futures Markets, Volume 14, Issue 2.[15] Tsantekidis, A., Passalis, N., Tefas, A., Kanniainen, J., Gabbouj, M., & Iosifidis, A. (2017). Forecasting stock prices from the limit order book using convolutional neural networks. In 2017 IEEE 19th Conference on Business Informatics (CBI) (Vol. 1, pp. 7-12). IEEE.[16] 吳永樂， (2014)。運用支持向量機和決策樹預測台指期走勢，國立政治大學，碩士論文。[17] 賴嘉蔚，(2017)。卷積神經網路預測時間序列能力分析，國立政治大學，碩士論文。[18] 劉昭雨，(2017)。卷積神經網路在金融技術指標之應用，國立東華大學，碩士論文。 zh_TW dc.identifier.doi (DOI) 10.6814/NCCU201900103 en_US
