臺灣博碩士論文加值系統

機器學習為達到人工智慧之方法，在股票預測中，機器學習方法具有非線性逼近能力、自學性與複雜系統控制能力等優勢，為此本文使用前饋神經網路模型與支持向量機模型。AH溢價指數是同時於A股市場和H股市場上市的中國大陸公司的A股相對H股的溢價或折讓，既可反映A股市場也可反映H股市場的上市公司表現。本研究利用機器學習與計量經濟模型方法，以AH溢價指數漲跌為研究標的。AH溢價指數漲跌之預測實證結果顯示，機器學習方法比計量經濟模型方法預測準確度更高。根據1,465種模型組合對比分析顯示，多層前饋神經網路模型以70%準確率略高於支持向量機模型67%準確率與計量經濟模型57%準確率。此外，本文還針對回溯天數、數據選取、資料前置處理、隱藏層數、隱藏節點數、激活函數、損失函數、核函數對模型預測準確率影響進行探討。

Machine learning is an approach of artificial intelligence. It exhibits nonlinearity in approximation, self-learning and the capability in solving complex problems. The AH Premium Index is designed to reflect the absolute price premium (or discount) of AH Companies. This thesis aims to predict AH premium index’s rise or fall the next day with multilayer feedforward neural networks (MFNNs), support vector machines (SVMs), and econometric models. It varies activation functions, loss functions, kernel functions, moving window, hidden layers, input variables and data preprocessing for controlling purpose. Among 1,465 parameter combinations, the MFNN model is found to have the highest performance (70%) followed by the SVM model (67%) and the econometric model (57%). The result indicates that machine learning techniques are better than econometric models in predicting AH premium index’s rise or fall the next day.

誌謝 I
中文摘要 II
英文摘要 III
目錄 IV
圖目錄 VII
表目錄 VIII
CHAPTER 1 緒論 1
1.1 研究動機與目的 1
1.2 本文架構 3
CHAPTER 2 相關文獻回顧 4
2.1 A股、H股與AH溢價指數概述 4
2.2 股票預測相關文獻回顧 13
CHAPTER 3 預測方法相關文獻 15
3.1 前饋神經網路 15
3.1.1 激活函數 17
3.1.2 損失函數 18
3.2 支持向量機 19
3.2.1 核函數 19
3.3 計量經濟模型 20
CHAPTER 4 實驗設計 21
4.1 資料來源 21
4.2 資料預處理 21
4.3 模型概述 23
4.3.1 多層前饋神經網路模型 24
4.3.2 支持向量機模型 26
4.3.3 計量經濟模型 28
CHAPTER 5 實驗結果 30
5.1 最佳結果 30
5.2 前饋類神經網路 32
5.2.1 最佳隱藏節點數 33
5.2.2 隱藏層層數 34
5.2.3 激活函數 35
5.2.4 損失函數 35
5.2.5 資料選取 36
5.2.6 資料預處理之運算處理 37
5.2.7 滾動預測之回溯天數 38
5.3 支持向量機 39
5.3.1 核函數 39
5.3.2 資料選取與資料預處理 40
5.4 計量經濟模型 43
CHAPTER 6 結論 44
參考文獻 46
附錄 51

Altman, E. I., Marco, G., & Varetto, F. (1994). Corporate Distress Diagnosis: Comparisons Using Linear Discriminant Analysis and Neural Networks (the Italian Experience). Journal of Banking & Finance, 18(3), 505–529. https://doi.org/10.1016/0378-4266(94)90007-8
Bailey, W. (1994). Risk and Return on China’s New Stock Markets: Some Preliminary Evidence. Pacific-Basin Finance Journal, 2(2–3), 243–260. https://doi.org/10.1016/0927-538X(94)90019-1
Budhani, N., Jha, C. K., & Budhani, S. K. (2014). Prediction of Stock Market Using Artificial Neural Network. In 2014 International Conference of Soft Computing Techniques for Engineering and Technology (ICSCTET) (pp. 1–8). IEEE. https://doi.org/10.1109/ICSCTET.2015.7371196
Burrell, P. R., & Folarin, B. O. (1997). The Impact of Neural Networks in Finance. Neural Computing & Applications, 6(4), 193–200. https://doi.org/10.1007/BF01501506
Chakravarty, S., Sarkar, A., & Wu, L. (1998). Information Asymmetry, Market Segmentation and the Pricing of Cross-listed Shares: Theory and Evidence From Chinese A and B Shares. Journal of International Financial Markets, Institutions and Money, 8(3–4), 325–356. https://doi.org/10.1016/S1042-4431(98)00041-9
Chowdhry, B., & Nanda, V. (1991). Multimarket Trading and Market Liquidity. The Review of Financial Studies, 4(3), 483–511.
Cybenko, G. (1992). Approximation by Superpositions of a Sigmoidal Function. Mathematics of Control, Signals, and Systems, 5(4), 455–455. https://doi.org/10.1007/BF02134016
Farias Nazário, R. T., e Silva, J. L., Sobreiro, V. A., & Kimura, H. (2017). A Literature Review of Technical Analysis on Stock Markets. The Quarterly Review of Economics and Finance, 66, 115–126. https://doi.org/10.1016/j.qref.2017.01.014
Fernald, J., & Rogers, J. H. (2002). Puzzles in the Chinese Stock Market. Review of Economics and Statistics, 84(3), 416–432. https://doi.org/10.1162/003465302320259448
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. Cambridge, MA : MIT Press, 2016.
Gulcehre, C., Moczulski, M., Denil, M., & Bengio, Y. (2016). Noisy Activation Functions. Retrieved from http://arxiv.org/abs/1603.00391
Hang Seng Indexes Company Limited. (2017a). Index Methodology for Managing the Hang Seng Stock Connect China AH Index Series. Retrieved from https://www.hsi.com.hk/static/uploads/contents/zh_hk/dl_centre/methodologies/IM_chinaahc.pdf
Hang Seng Indexes Company Limited. (2017b). Trading Mechanism of Closing Auction Session (CAS) in the Securities Market. Retrieved from https://www.hkex.com.hk/-/media/HKEX-Market/News/News-Release/2016/160720news/Trading-Mechanism-of-the-CAS-in-the-Securities-Market.pdf?la=en
Haykin, S. S. (2009). Neural Networks and Learning Machines. New York: Prentice Hall.
Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer Feedforward Networks are Universal Approximators. Neural Networks, 2(5), 359–366. https://doi.org/10.1016/0893-6080(89)90020-8
Hsu, C.-W., Chang, C.-C., & Lin, C.-J. (2003). A Practical Guide to Support Vector Classification. Http://Www.Csie.Ntu.Edu.Tw/~cjlin/Papers/Guide/Guide.Pdf.
Lin, H.-T., & Lin, C.-J. (2005). A Study on Sigmoid Kernels for SVM and the Training of non-psd. Http://Www.Csie.Ntu.Edu.Tw/~cjlin/Papers/Tanh.Pdf. https://doi.org/20060927122853789664
Ma, X. (1996). Capital Controls, Market Segmentation and Stock Prices: Evidence From the Chinese Stock Market. Pacific-Basin Finance Journal, 4(2–3), 219–239. https://doi.org/10.1016/0927-538X(96)00012-1
Mercer, J. (1909). Functions of Positive and Negative Type, and Their Connection with the Theory of Integral Equations. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 83(559), 69–70. https://doi.org/10.1098/rspa.1909.0075
Nguyen, D. H., & Widrow, B. (1991). Neural Networks for Self-Learning Control Systems. International Journal of Control, 54(6), 1439–1451. https://doi.org/10.1080/00207179108934220
Nicholas Refenes, A., Zapranis, A., & Francis, G. (1994). Stock Performance Modeling Using Neural Networks: A Comparative Study with Regression Models. Neural Networks, 7(2), 375–388. https://doi.org/10.1016/0893-6080(94)90030-2
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1985). Learning Internal Representations by Error Propagation. In D.E. Rumelhart and J. L. McClelland (Ed.), Parallel Distributed Processing (pp. 318–362). Cambridge, Mass. : MIT Press.
Schierholt, K., & Dagli, C. H. (1996). Stock Market Prediction Using Different Neural Network Classification Architectures. In IEEE/IAFE 1996 Conference on Computational Intelligence for Financial Engineering (CIFEr) (pp. 72–78). IEEE. https://doi.org/10.1109/CIFER.1996.501826
Stulz, R. M., & Wasserfallen, W. (1995). Foreign Equity Investment Restrictions, Capital Flight, and Shareholder Wealth Maximization Theory and Evidence. Review of Financial Studies, 8(4), 1019–1057.
Sun, Q., & Tong, W. H. S. (2000). The Effect of Market Segmentation on Stock Prices: The China Syndrome. Journal of Banking & Finance, 24(12), 1875–1902. https://doi.org/10.1016/S0378-4266(99)00121-1
Tsang, P. M., Kwok, P., Choy, S. O., Kwan, R., Ng, S. C., Mak, J., … Wong, T.-L. (2007). Design and Implementation of NN5 for Hong Kong Stock Price Forecasting. Engineering Applications of Artificial Intelligence, 20(4), 453–461. https://doi.org/10.1016/j.engappai.2006.10.002
Wang, S. S., & Jiang, L. (2004). Location of Trade, Ownership Restrictions, and Market Illiquidity: Examining Chinese A- and H-shares. Journal of Banking & Finance, 28(6), 1273–1297. https://doi.org/10.1016/S0378-4266(03)00119-5
World Bank. (2018). Gross Domestic Product 2017. Retrieved from http://databank. worldbank.org/data/download/GDP.pdf
Wu, M., & Yu, Q. (2010). Empirical Analysis of the Impact Factors of the A Shares and H Shares of the Price Differences. In 2010 International Conference on E-Product E-Service and E-Entertainment (pp. 1–4). IEEE. https://doi.org/10.1109/ICEEE.2010.5661626
Yoon, Y., Swales, G., & Margavio, T. M. (1993). A Comparison of Discriminant Analysis Versus Artificial Neural Networks. Journal of the Operational Research Society, 44(1), 51–60. https://doi.org/10.1057/jors.1993.6
Zhang, G. P. (2004). Forecasting Stock Returns with Artificial Neural Networks. In Neural Networks in Business Forecasting (pp. 47–79). Hershey, PA: Idea Group.
上海證券交易所. (2018). 2018年市場資料. Retrieved from http://www.sse.com.cn/aboutus/publication/factbook/documents/c/4563235.pdf
台灣經濟新報. (2014). 中國市場簡介. Retrieved from https://www.tej.com.tw/twsite/TEJWeb/tw/database/doc/cn.pdf
呂欣. (2010). 基於神經網絡股票價格預測模型及系統的研究. 吉林大學(碩士論文).
姚培福. (2007). 人工神經網路在股票預測中的應用與研究. 昆明理工大學(碩士論文).
孫丹、 張秀豔. (2002). 基於人工神經網絡的股市預測模型. 吉林大學學報, 20(4), 68–70.
師智斌、陳立潮、靳雁霞. (2003). 基於神經網絡的股票交易數據的預測研究. 華北工學院學報, 34(6), 12–16.
張吉剛、梁娜. (2008). 基於SOM網絡－主成分－BP網絡的股價預測. 統計與決策, 2008(6), 158–160.
徐壽福. (2009). 「雙重上市」公司A、H股價格差異的因素研究. 證券市場導報, 2009(2), 54–60.
範海燕. (2009). 市場分割下A股和H股的價格差異研究. 山東大學(碩士論文).
葉怡成. (1995). 類神經網路: 模式應用與實作. 臺北市: 儒林.
邱振祥. (2010). 中國A股與香港H股之折溢價實證分析. 國立臺灣大學（碩士論文）.
閻威武、常俊林、邵惠鶴. (2004). 基於滾動時間窗的最小二乘支持向量機回歸估計方法及仿真. 上海交通大學學報, 38(004), 524–526.
陳永義、熊秋芬. (2011). 支持向量機方法應用教程. (陳永義、熊秋芬, Eds.). 北京: 氣象出版社.
韓德宗. (2006). A股和H股市場軟分割因素研究－兼論推出QDⅡ的步驟和時機. 商業經濟與管理, 2006(3), 42–46.
香港交易所. (2014). 概念文件：不同投票權架構. Retrieved from http://www.hkex.com.hk/-/media/HKEX-Market/News/Market-Consultations/2011-to-2015/August-2014-Weighted-Voting-Rights/Consultation-paper/cp2014082_c.pdf
香港交易所. (2018a). 投資者資料文件. Retrieved from http://www.hkex.com.hk/-/media/HKEX-Market/Mutual-Market/Stock-Connect/Getting-Started/Information-Booklet-and-FAQ/Information-Book-for-Investors/Investor_Book_Cn_(May).pdf
香港交易所. (2018b). 綜合主板上市規則之聲明. Retrieved from http://www.hkex.com.hk/-/media/HKEX-Market/Listing/Rules-and-Guidance/Listing-Rules/Consolidated-PDFs/Main-Board-Listing-Rules/consol_mb_tc.pdf?la=zh-HK
香港交易所. (2018c). 香港交易所市場資料 2017. Retrieved from http://www.hkex.com.hk/Market-Data/Statistics/Consolidated-Reports/HKEX-Fact-Book?sc_lang=zh-HK
高大啟. (1998). 有教師的線性基本函數前向三層神經網絡結構研究. 計算機學報, 21(1), 80–86.