1. NCCU Academic Hub
  2. 學術產出
  3. College of Commerce
  4. Theses
  5. Item

學術產出-Theses

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

題名 以循環生成對抗網路預測股價量能動態關係
Using Cycle GAN to predict dynamic correlation between price and volume
作者 林奕廷
貢獻者 姜國輝
Chiang, Johannes K.
林奕廷
關鍵詞 量價關係
系統動態學
循環生成對抗網路
技術分析
日期 2019
上傳時間 7-Aug-2019 16:08:54 (UTC+8)
摘要 預測股價漲跌幅是投資人需要的資訊。技術分析方法種類眾多，當中量價關係為其主流方法之一。本研究使用2007年至 2017年 台積電股票資料，運用首次提出的方法來觀察量價關係，透過循環生成對抗網路(Cycle GAN)結合卷積網路(Convolution Neural Network)與殘差網路(Residual Neural Network)學習量價間的聯合(joint)作用，並得出潛在量價。並參考系統動態學(system dynamic)，將潛在量價與目前量價的位能差距作為影響質量的勢能，透過神經網路轉換成推力， 市值、稅金、昨日漲跌幅視為質量、摩擦力、彈簧力，模擬現實股價上漲下跌的因素。為了方便使用者使用，本研究再將預測股價結合布林帶(Bollinger Bands，BBands)，透過其延伸指標%b指標(Percent b)決定交易信號。經研究證實，結合股價預測的布林帶，比僅有布林帶的平均投資報酬率提升了30%左右。
The increasing demands for stock market prediction plays an important role in nowadays. Via Price -Volume Relation, we are able to access more insight for investors. We propose a novel generative model based on cyclic-consistent generative adversarial network (CycleGAN), which consists of Convolution Neural Network(CNN) and Residual Network(RESNET) for stock market prediction, in particular, Price-Volume Relation. In the reference to system dynamic, we can simulate potential fluctuation of current stock market via Neural network. Considering all the factors of stock market fluctuation e.g. value of stocks, tax, and current price changes as massive, friction, and spring force in system dynamic, our model is able to simulate the fluctuation and gain the ideal forecast. Our research is based on 2007-2018 TSMC stock dataset. For user friendly purpose, we compose BBands (Bollinger Bands) and stock price prediction model to use its derived indicator , %b indicator, to make trade signal. With BBands and our stock price prediction model, its average ROI(Return on investment) has increased 30% efficiency, which is the better result with merely BBands base on our experiment.
參考文獻 1. 胡依淳，民107，深度卷積神經網路中卷積層之分析及比較。國立暨南國際大學電機工程學系碩士論文。
2. 高士軒，民97，「價量關係：量是否為價格發現的先行指標」，逢甲大學財務金融學所碩士論文。
3. 陳盈臻，民102。台灣股票市場量價背離之實證研究，佛光大學管理學系碩士論文。
4. Ahmed, A. S., Schneible, R. A. J., and Stevens, D. E. 2003. An Empirical Analysis of the Effect of Online Trading on Stock Price and Trading Volume Reactions to Earnings Announcements. Contemporary Accounting Research, 20, 413-439.
5. Crouch, R. L. 1970. The volume of transactions and price changes on the New York
6. DeMark, T. R. 1984. The New Science of Technical analysis. New York： John Wiley and Sons, Inc.
7. Gers, F. A., Schmidhuber, J., & Cummins, F. 1999. Learning to forget： Continual prediction with LSTM.
8. Goodfellow, Ian, et al. 2014. "Generative adversarial nets." Advances in neural information processing systems.
9. Granger, C. W. J., and Morgenstern, O. 1963. Spectral Analysis of New York Stock Market Prices. Kyklos, 16, 1-27.
10. He, Kaiming, et al. 2016."Deep residual learning for image recognition." Proceedings of the IEEE conference on computer vision and pattern recognition.
11. Seely, Samuel. 1972. An Introduction to Engineering Systems.
12. Sheu, Her-Jiun, Wu, Soushan and Ku, Kuang-Ping 1998. The cross-sectional relationships between stock returns and market beta, trading volume sales-to-price in Taiwan. International Review of Financial Analysis, 7, 1-18.
13. Ying, C. C. (1966). Stock Market Prices and Volumes of Sales. Econometrica, 34, 676-685.
14. Zhang, Liheng, Charu Aggarwal, and Guo-Jun Qi. 2017. "Stock price prediction via discovering multi-frequency trading patterns." Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.
15. Zhu, Jun-Yan, et al. 2017. "Unpaired image-to-image translation using cycle-consistent adversarial networks." Proceedings of the IEEE International Conference on Computer Vision.
16. 網頁資料 Jon Bruner “Generative Adversarial Networks for Beginners”(https：//github.com/jonbruner/generative-adversarial-networks/blob/master/gan-notebook.ipynb)
17. 網頁資料 ADAM HAYES “Technical Analysis Definition”(https：//www.investopedia.com/terms/t/technicalanalysis.asp)
18. 網頁資料 “布林帶”(https：//zh.wikipedia.org/wiki/%E5%B8%83%E6%9E%97%E5%B8%A6)
描述 碩士
國立政治大學
資訊管理學系
106356039
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0106356039
資料類型 thesis
dc.contributor.advisor 姜國輝zh_TW
dc.contributor.advisor Chiang, Johannes K.en_US
dc.contributor.author (Authors) 林奕廷zh_TW
dc.creator (作者) 林奕廷zh_TW
dc.date (日期) 2019en_US
dc.date.accessioned 7-Aug-2019 16:08:54 (UTC+8)-
dc.date.available 7-Aug-2019 16:08:54 (UTC+8)-
dc.date.issued (上傳時間) 7-Aug-2019 16:08:54 (UTC+8)-
dc.identifier (Other Identifiers) G0106356039en_US
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/124720-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 資訊管理學系zh_TW
dc.description (描述) 106356039zh_TW
dc.description.abstract (摘要) 預測股價漲跌幅是投資人需要的資訊。技術分析方法種類眾多，當中量價關係為其主流方法之一。本研究使用2007年至 2017年 台積電股票資料，運用首次提出的方法來觀察量價關係，透過循環生成對抗網路(Cycle GAN)結合卷積網路(Convolution Neural Network)與殘差網路(Residual Neural Network)學習量價間的聯合(joint)作用，並得出潛在量價。並參考系統動態學(system dynamic)，將潛在量價與目前量價的位能差距作為影響質量的勢能，透過神經網路轉換成推力， 市值、稅金、昨日漲跌幅視為質量、摩擦力、彈簧力，模擬現實股價上漲下跌的因素。為了方便使用者使用，本研究再將預測股價結合布林帶(Bollinger Bands，BBands)，透過其延伸指標%b指標(Percent b)決定交易信號。經研究證實，結合股價預測的布林帶，比僅有布林帶的平均投資報酬率提升了30%左右。zh_TW
dc.description.abstract (摘要) The increasing demands for stock market prediction plays an important role in nowadays. Via Price -Volume Relation, we are able to access more insight for investors. We propose a novel generative model based on cyclic-consistent generative adversarial network (CycleGAN), which consists of Convolution Neural Network(CNN) and Residual Network(RESNET) for stock market prediction, in particular, Price-Volume Relation. In the reference to system dynamic, we can simulate potential fluctuation of current stock market via Neural network. Considering all the factors of stock market fluctuation e.g. value of stocks, tax, and current price changes as massive, friction, and spring force in system dynamic, our model is able to simulate the fluctuation and gain the ideal forecast. Our research is based on 2007-2018 TSMC stock dataset. For user friendly purpose, we compose BBands (Bollinger Bands) and stock price prediction model to use its derived indicator , %b indicator, to make trade signal. With BBands and our stock price prediction model, its average ROI(Return on investment) has increased 30% efficiency, which is the better result with merely BBands base on our experiment.en_US
dc.description.tableofcontents 第一章 緒論 1
第一節 研究背景 1
第二節 研究動機 2
第三節 研究目的 3
第二章 文獻探討 4
第一節 生成對抗網路(Generative Adversarial Network , GAN) 4
第三節 卷積神經網絡(Convolutional Neural Network , CNN) 5
第四節 殘差網路 (Residual Neural Network , ResNet ) 6
第五節 Long Short-Term Memory (LSTM) 6
第六節 量價關係 7
1. 量價之間的關係 7
2. 股價與成交量有聯合作用 7
第七節 系統動態學 8
1. Mass (質量) 8
2. Damper(摩擦力) 8
3. Spring(彈簧) 8
第八節 布林帶 9
1. 布林帶定義 9
2. 延伸指標—%b指標 10
第三章 研究方法 11
第一節 資料蒐集 11
第二節 資料前處理 11
1. 股價與交易量正規化 12
2. 直接正規化 12
3. 訓練/驗證/測試資料切割 13
第三節 神經網路設計 13
1. 循環生成對抗網路 設計 13
2. 模擬系統動態學預測股價 15
3. 無模擬系統動態學預測股價 16
4. 神經網路設計 16
A. 運用 RESNET建構神經網路 16
B. 運用LSTM建構神經網路 17
5. 布林帶設計 18
第四章 實驗結果 19
第一節 模型評估 19
1. MSE(Mean Squared Error) 19
2. MAE(Mean Abs Error) 19
3. 投資報酬率(Return On Investment，ROI) 19
第二節 環生成對抗網路學習量價關係成果 19
第三節 預測股價結論 21
1. 正規化與無正規化 21
2. 預測股價表現 21
第四節 預測交易信號與回報率 22
1. 布林帶 22
2. 布林帶與模擬系統動態學LSTM預測股價整合 23
3. 布林帶與模擬系統動態學RESNET預測股價整合 24
4. 布林帶與無模擬系統動態學LSTM預測股價整合 25
5. 布林帶與無模擬系統動態學RESNET預測股價整合 26
6. 投資報酬率結論 28
第五章 研究結論與建議 29
第一節 結論 29
第二節 未來建議 29
參考文獻 30
附錄 32		zh_TW
dc.format.extent 11167589 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0106356039en_US
dc.subject (關鍵詞) 量價關係zh_TW
dc.subject (關鍵詞) 系統動態學zh_TW
dc.subject (關鍵詞) 循環生成對抗網路zh_TW
dc.subject (關鍵詞) 技術分析zh_TW
dc.title (題名) 以循環生成對抗網路預測股價量能動態關係zh_TW
dc.title (題名) Using Cycle GAN to predict dynamic correlation between price and volumeen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) 1. 胡依淳，民107，深度卷積神經網路中卷積層之分析及比較。國立暨南國際大學電機工程學系碩士論文。
2. 高士軒，民97，「價量關係：量是否為價格發現的先行指標」，逢甲大學財務金融學所碩士論文。
3. 陳盈臻，民102。台灣股票市場量價背離之實證研究，佛光大學管理學系碩士論文。
4. Ahmed, A. S., Schneible, R. A. J., and Stevens, D. E. 2003. An Empirical Analysis of the Effect of Online Trading on Stock Price and Trading Volume Reactions to Earnings Announcements. Contemporary Accounting Research, 20, 413-439.
5. Crouch, R. L. 1970. The volume of transactions and price changes on the New York
6. DeMark, T. R. 1984. The New Science of Technical analysis. New York： John Wiley and Sons, Inc.
7. Gers, F. A., Schmidhuber, J., & Cummins, F. 1999. Learning to forget： Continual prediction with LSTM.
8. Goodfellow, Ian, et al. 2014. "Generative adversarial nets." Advances in neural information processing systems.
9. Granger, C. W. J., and Morgenstern, O. 1963. Spectral Analysis of New York Stock Market Prices. Kyklos, 16, 1-27.
10. He, Kaiming, et al. 2016."Deep residual learning for image recognition." Proceedings of the IEEE conference on computer vision and pattern recognition.
11. Seely, Samuel. 1972. An Introduction to Engineering Systems.
12. Sheu, Her-Jiun, Wu, Soushan and Ku, Kuang-Ping 1998. The cross-sectional relationships between stock returns and market beta, trading volume sales-to-price in Taiwan. International Review of Financial Analysis, 7, 1-18.
13. Ying, C. C. (1966). Stock Market Prices and Volumes of Sales. Econometrica, 34, 676-685.
14. Zhang, Liheng, Charu Aggarwal, and Guo-Jun Qi. 2017. "Stock price prediction via discovering multi-frequency trading patterns." Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.
15. Zhu, Jun-Yan, et al. 2017. "Unpaired image-to-image translation using cycle-consistent adversarial networks." Proceedings of the IEEE International Conference on Computer Vision.
16. 網頁資料 Jon Bruner “Generative Adversarial Networks for Beginners”(https：//github.com/jonbruner/generative-adversarial-networks/blob/master/gan-notebook.ipynb)
17. 網頁資料 ADAM HAYES “Technical Analysis Definition”(https：//www.investopedia.com/terms/t/technicalanalysis.asp)
18. 網頁資料 “布林帶”(https：//zh.wikipedia.org/wiki/%E5%B8%83%E6%9E%97%E5%B8%A6)		zh_TW
dc.identifier.doi (DOI) 10.6814/NCCU201900551en_US