臺灣博碩士論文加值系統

台灣股票市場成立迄今已四十六年，股票投資已成為一般民眾主要的理財工具，然而影響股價波動的因素眾多，包含總體經濟、國際股匯市、政治面、公司基本面、技術面、籌碼面、消息面等，市場上的投資大眾為了有效預測股價的走勢，遂有許多分析方法乃因應而生。基本上，這些方法主要可區分成「基本分析」或「技術方析」兩大類。而所謂的基本分析主要是以經濟、產業及公司營運等面向的資料為基礎進行股票價格分析；技術分析則是透過整體市場及個別股票資料的價格及成交量為基礎建構不同技術指標進行分析。在技術分析的眾多技術指標中，由於K線圖 (K-chart) 可以簡單的圖形化方式反應市場的趨勢且投資人可以藉由K線型態的成立與否來判斷市場未來走向，所以常被投資大眾所採用。
事實上，在學術上已有許多的研究是根據K線型態使用數值運算判斷型態轉折進行分析的。此外，有許多學者也嘗試應用資料分群(clustering)的概念，以自組織映射圖網路 (Self-organizing map, SOM) 建立K線型態分群模式，試圖找出指數資料中所有可能出現的K線型態，進行走勢的判斷。但是由於傳統的SOM模式在模式訓練前必須先給定拓撲網路 (Topology map) 的大小，所以在訓練過程中並無法根據資料的特性來增加單元 (Unit) 數。再者，SOM模式由於受限於二度空間拓撲座標之假設，亦無法適當呈現具有層級性的資料型態。換言之，自組織映射圖網路並無法有效地處理具有階層特性的指數資料。
在本研究中，我們針對股價加權指數及個股股價資訊，採用Dittenbach et al. (2000) 所提出之成長型階層式自組織映射圖網路 (Growing hierarchical self-organizing map, GHSOM) 技術進行K線型態分群模式的建構。藉由所建構之分群模式，成功的尋找出隱藏在累積資料中的K線型態及正確的資料層級特性。此外，我們也根據分群模式的分析結果採用分類迴歸樹(Classification and Regression tree, CART)的方法，進行即時指數資料的監控，以預測市場未來趨勢。

Because of low interest rates in recent years, the stock market has been a very popular financial investment channel. However, financial investment is a knowledge-intensive industry. Without sufficient understanding, it is hard for any investor to get profit from it. In the past decade, with the advances in electronic transactions, vast amounts of data have been collected. Thus, the emergence of knowledge discovery technology enables the building up a financial investment decision support system.
There are various techniques of knowledge discovery have been employed in the stock market. Basically, they can be divided into two parts, technical analysis and forecasting of stock time series data. The former applies the charting heuristics of technical analysis to identify the bull flag by template match and to establish trading rules. The latter focuses on the forecast of stock prices, which often employs a statistical or artificial intelligence (AI) approach to facilitate the trading strategy-making.
In this thesis, we propose a hybrid approach on the basis of the knowledge discovery methodology by using Growing hierarchical self-organizing map (GHSOM) model, integrating K-chart technical analysis for feature representation of stock price movements. Moreover, the possibility of integrated GHSOM and Classification and regression tree (CART) methodology is examed. The investigated results showthat the application of GHSOM and CART in data mining provides the efficient identification and classification methods for financial investment data. And in terms of the successful identification of the relationship within data, the better prediction modeling can be found.

摘　要 i
Abstract ii
誌　謝 iii
目　錄 iv
表目錄 v
圖目錄 vi
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 4
1.3 研究架構 4
第二章 文獻探討 6
2.1 類神經網路 6
2.1.1 輻射基底函數類神經網路 8
2.2 分群方法 9
2.2.1 自組織網路映射圖 9
2.2.2 成長型階層式自組織網路映射圖 11
2.3 分類迴歸樹 12
第三章 研究方法 13
3.1 自組織網路映射圖 13
3.2 成長型階層式自組織網路映射圖 15
3.3 輻射基底函數類神經網路 19
3.4 分類迴歸樹 22
第四章 研究結果 24
4.1 資料描述 24
4.2 資料前處理 24
4.2.1 資料區段與切割視窗 24
4.2.2 標準化 26
4.3 實證結果 27
4.4 模式之比較 30
第五章 結論與建議 32
5.1 研究結論 32
5.2 未來研究方向 32
參考文獻 33
附錄 38
附錄A：GHSOM各群圖形趨勢 38
附錄B：CART最佳樹狀圖-節點編號 39
附錄C：CART最佳樹狀規則 40

[1] 杜金龍，基本分析在台灣股市應用的訣竅，台北：財訊，2002
[2] 吳榮訓、廖高賢，「使用基因程式規劃預測股票買賣時機」，資訊管理學報，第十四卷，第二期，2007，第111-138頁。
[3] 臺灣內政部統計處。取自：http://www.moi.gov.tw/stat/
[4] 臺灣行政院金融監督管理委員會證券期貨局。取自：http://www.sfb.gov.tw/
[5] A. Rauber, E. Pampalk and D. Merkl, “Using PsychoAcoustic Models and Self Organizing Maps to Create a Hierarchical Structuring of Music by Sound Similarity,” Proceedings of the 3rd International Symposium on Music Information Retrieval, 2002, pp. 71-80.
[6] A. Soriano-Asensi, J. D. Marti´n-Guerrero, E. Soria-Olivas, A. Palomares, R. Magdalena-Benedito and A. J. Serrano-Lo´pez, “Web mining based on Growing Hierarchical Self-Organizing Maps: Analysis of a real citizen web portal,” Expert Systems with Applications, vol. 34, no. 4, 2008, pp. 2988-2994.
[7] A. Vellido, P. J. G. Lisboa and J. Vaughan, “Neural networks in business: A survey of applications (1992-1998),” Expert Systems with Applications, vol. 17, no. 1, 1999, pp. 51-70.
[8] B. Bai, J. Yen and X. Yang, “False financial statements: Characteristics of china’s listed companies and CART detecting approach”, International Journal of Information Technology & Decision Making, vol. 7, no. 2, 2008, pp. 339-359
[9] B. Fritzke, “Growing grid－a self-organizing network with constant neighborhood range and adaption strength,” Neural Processing Letter, vol. 2, no. 5, 1995, pp. 1-5.
[10] C.-C. Hsu, ”Generalizing Self-Organizing Map for Categorical Data”, IEEE Transactions on Neural Networks, vol. 17, no. 2, 2006, pp. 294-304.
[11] C. Ohmann, V. Moustakis, Q. Yang and K. Lang, “Evaluation of automatic knowledge acquisition techniques in the diagnosis of acute abdominal pain,” Artificial Intelligence in Medicine, vol. 8, no. 1, 1996, pp. 23-36.
[12] C. R. Cañizares and V. Pilipovik, “The use of radial basis function networks (RBFN) to predict critical waterparameters in desalination plants,” Expert Systems with Applications, 2010.
[13] C. Y. Tsao and S. H. Chen, “Self-organizing maps as a foundation for charting or geometric pattern recognition in financial time series,” Proceedings of the IEEE International Conference on Computational Intelligence for Financial Engineering, Hong Kong, 2003.
[14] D. G. Champernowne, “Sampling theory applied to autoregressive schemes,” Journal of the Royal Statistical Society: Series B, vol. 10, 1948, pp. 204-231.
[15] D. J. Hand, Data Mining: Statistics and More?, The American Statistician, 1998, vol. 52.
[16] D. S. Broomhead and D. Lowe, “Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks,” Complex Syst., vol. 2, 1988, pp. 321-355.
[17] F. R. Lin and C. H. Liang, “Storyline-based summarization for news topic retrospection,” Decision Support Systems, vol. 45, no. 3, 2008, pp.473-490.
[18] G. E. P. Box and G. M. Jenkins, Time series analysis: Forecasting and control, 3rd Edition, Prentice Hall, Englewood Cliffs, 1994.
[19] G. V. Kass, “An Exploratory Technique for Investigating Large Quantities of Categorical Data,” Applied Statistics, vol. 29, no. 2, 1980, pp. 119-127.
[20] H. Curt, “The devil’s in the detail: techniques, tools, and applications for database mining and knowledge discovery-part 1,” Intelligent Software Strategies, vol.6, 1995, pp. 1-15.
[21] H. Zeng, L. Chen, X. Han, X. Zhao and Y. Ma, “Classification and regression tree (CART) for analysis of soybean yield variability among fields in Northeast China: The importance of phosphorus application rates under drought conditions,” Agriculture, Ecosystems and Environment, vol. 132, 2009, pp. 98-105.
[22] J. A. Morgan and J. N. Sonquist, “Problems in the analysis of drug data and a proposal,” Journal of the American Statistical Association, vol. 58 , 1963, pp. 415-434.
[23] J. R. Quinlan, “Induction of decision trees,” Machine Learning, vol. 1, no. 1, 1986, pp. 81-106.
[24] J.-Y. Shih, Y.-J. Chang and W.-H. Chen, “Using GHSOM to construct legal maps for Taiwan’s securities and futures markets,” Expert Systems with Applications, vol. 34, 2008, pp. 850-858.
[25] K. H. Lee and G. S. Jo, “Expert system for predicting stock market timing using a candlestick chart,” Expert Systems with Applications, vol. 16, no. 4, 1999, pp. 357-364.
[26] L. Breiman, J. H. Friedman, R. A. Olshen and C. J. Stone, Classification and Regression Trees, Monterey, Ca: Wadsworth, 1984.
[27] L.-C. Chang, F.-J. Chang and Y.-P. Wang, “Auto-configuring radial basis function networks for chaotic time series and flood forecasting”, Hydrol. Process, vol. 23, no. 17, 2009, pp. 2450-2459.
[28] M. Bevilacqua, M. Braglia and R. Montanari, “The classification and regression tree approach to pump failure rate analysis,” Reliability Engineering and System Safety, vol. 79, 2003, pp. 59-67.
[29] M. Boudour, and A. Hellal, “The growing hierarchical self-organizing feature maps and genetic algorithms for large scale power system security,” Proceedings of the International Conference on Adaptive and Natural Computing Algorithms, 2005, pp. 159-163.
[30] M. Dittenbach, A. Rauber and D. Merkl, “Uncovering hierarchical structure in data using the growing hierarchical self-organizing map,” Neurocomputing, vol. 48, 2002, pp. 199-216.
[31] M. Dittenbach, D. Merkl and A. Rauber, “The Growing Hierarchical Self-Organizing Map,” Proceedings of the International Joint Conference on Neural Networks, vol. 6, 2000, pp. 15-19
[32] M. J. D. Powell, “Radial basis functions for multivariable interpolation: A review,” In Algorithms for Approximation, J. C. Mason and M. G. Cox, Eds. Oxford, 1987, pp. 143-167.
[33] M. Lehtokangas, J. Saarinen and K. Kaski, ”A Network of autogressive processing units for time series Modeling.”, Applied Mathematics and Computation, vol. 75, 1996, pp. 151-165.
[34] M. W. Carven and J. W. Shavlik, “Using neural networks for data mining,” Future Generation Computer Systems, vol. 13, 1997, pp. 211-229.
[35] N. Pears, T. Heseltine and M. Romero, “From 3D point clouds to pose-normalised depth maps,” International Journal of Computer Vision, pp. 1-25.
[36] Q. Zhu, T. Zhang and S. Fei, “Adaptive tracking control for input delayed MIMO nonlinear systems,” Neurocomputing, 2010.
[37] R. D. Edwards, J. Magee and W. H. C. Bassetti, Technical Analysis of Stock Trends, New York: CRC Press, 2001.
[38] R. F. Engle, “Autoregressive conditional heteroscedasticity with estimates of the variance of united kingdom inflation,” Econometrica, vol. 50, no. 4, 1982, pp. 987-1008.
[39] R. L. Hardy, “Multiquadric equations of topography and other irregular surfaces,” J. Geophys Res., vol. 76, no. 8, 1971, pp. 1905-1915.
[40] S. Haykin, Neural Networks: a Comprehensive Foundation, New Jersey: Prentice-Hall, Englewood Cliffs, 1999.
[41] S. Slušný, R. Neruda and P. Vidnerová, “Comparison of behavior-based and planning techniques on the small robot maze exploration problem,” Neural Networks, vol. 23, no. 4, 2010, pp. 560-567.
[42] S. V. Verdú, M. O. García, C. Senabre, A. G. Marín and F. J. G. Franco, “Classification, Filtering, and Identification of Electrical Customer Load Patterns Through the Use of Self-Organizing Maps,” IEEE Transactions on Power Systems, vol. 21, no. 4 , 2006, pp. 1672-1682.
[43] S. Y. Liu, L. Lu, G. G. Liao and J. P. Xuan, “Pattern discovery from time series using growing hierarchical self-organizing map,” Proceedings of the International Conference on Neural Information Processing, Hong Kong, 2006.
[44] T. Bollerslev, “Generalized autoregressive conditional heteroscedasticity,” Journal of Econometrics, vol. 31, no. 3, 1986, pp. 307-327.
[45] T. Kohonen, Self Organization and Associative Memory (3nd ed.). Berlin: Springer Verlag, 1989.
[46] W. Loh and N. Vanichsetakul, “Tree-structured classification via generalized discriminant analysis,” Journal of the American Statistical Association, vol. 83, 1988, pp. 715-728