支援向量機與支援向量迴歸於財務時間序列預測之應用

Financial Time Series Forecasting Using Support Vector Machine and Support Vector Regression

10.6338/JDA.200904_4(2).0003

呂奇傑(Chi-Jie Lu)；李天行(Tian-Shyug Lee)；高人龍(Jen-Lung Kao)；黃敏菁(Min-Ching Huang)

財務時間序列預測 ； 支援向量機 ； 支援向量迴歸 ； 類神經網路 ； Financial Time Series Forecasting ； Support Vector Machine ； Support Vector Regression ； Artificial Neural Network ； Taiwan Stock Weighted Exchange Index

Journal of Data Analysis

4卷2期（2009 / 04 / 01）

35 - 56

繁體中文

財務時間序列預測一直是各領域人才躍躍欲試的題材，然而因財務時間序列資料中的雜訊（Noise）、非定態（Non-stationary）及確定性混亂現象等特性的影響，提高了建構準確的財務時間序列預測模式之困難度。綜觀國內外文獻，由於傳統的預測模型常需符合模型特定的統計假設，限制了其實用性，因此不需過多假設及擁有學習能力特性的人工智慧（Artificial intelligence）方法便逐漸受到重視及使用，其中類神經網路（Artificial neural network, ANN）已成功應用在建構財務時間序列之預測模式上，但ANN仍存在著需要大量控制參數與容易落入區間極值（Local extreme）等問題。支援向量機（Support vector machine, SVM）與支援向量迴歸（Support vector regression, SVR）是一個以統計學習理論（Statistical learning theory）為基礎的預測方法，由於具有全域最佳解（Global optimum）與考慮結構風險（Structural risk）的特性，已成功應用在文字分類、影像識別、生物資訊等領域。本研究將嘗試以SVM與SVR為預測工具，建構財務時間序列預測模式，並且為驗證其有效性，將以預測台灣加權股價指數（TAIEX）之漲跌方向進行實證研究，所得之預測結果將與ANN中的倒傳遞類神經網路（Back-propagation neural network, BPN）的預測結果做比較。實證結果顯示，不論在漲跌幅度的預測誤差或是趨勢預測準確度的表現上，SVM與SVR均較BPN為佳，代表透過支援向量機與支援向量迴歸建構財務時間序列預測模式具有效性，能提供更佳的投資建議。

As financial time series are inherently noisy, non-stationary and deterministically chaotic, it is one of the most challenging applications of modern time series forecasting. Due to the advantages of building nonparametric and nonlinear models, artificial neural network (ANN) has also been successfully applied in time series prediction, especially in modeling financial time series forecasting problems. Support vector machine (SVM) and support vector regression (SVR) are novel neural network techniques based on statistical learning theory and structural risk minimization principle. They have been used in wide range of applications. The aims of this paper are to examine the ability of SVM and SVR in predicting the next trading day's vibration and direction of the Taiwan stock weighted exchange index (TAIEX). The forecasting results of TAIEX using back-propagation neural network (BPN) are used as benchmark for evaluating the performance of SVM and SVR. The experimental results show that SVM and SVR outperform the BPN model in forecasting performance.

1. Anderson, J. A.,Rosenfeld, E.(1998).Neurocomputing: Foundation of Research.Cambridge, MA:MIT Press.
2. Apergis, N.(2004).Inflation, output growth, volatility and causality: evidence from panel data and the G7 countries.Economics Letters,83,185-191.
3. Balachandher, K. G.,Fauzias, M. N.,Lai, M. M.(2002).An examination of the random walk model and technique trading rules in the Malaysian stock market.Quarterly Journal of Business & Economics,41,81-104.
4. Burbidge, R.,Trotter, M.,Buxton, B.,Holden, S.(2001).Drug design by machines learning: support vector machines for pharmaceutical data analysis.Computer & Chemistry,26,5-14.
5. Chang, C. C.,Lin, C. J.(2001).,未出版
6. Chen, A. S.,Leung, M. T.,Daouk, H.(2003).Application of neural networks to an emerging financial market: forecasting and trading the Taiwan Stock Index.Computers & Operations Research,30,901-923.
7. Cherkassky, V.,Ma, Y.(2004).Practical selection of SVM parameters and noise estimation for SVM regression.Neural Networks,17,113-126.
8. Cherkassky, V.,Mulier, F.(1999).Vapnik-Chervonenkis (VC) learning theory and its applications.IEEE Transaction on Neural Networks,10,985-987.
9. Cover, J. P.,Huang, C. J.(2003).The correlation between shocks to output and the price level: evidence from a multivariate GARCH model.Southern Economic Journal,70,75-92.
10. Deboeck, G. J.(1994).Trading on the Edge: Neural, Genetic and Fuzzy Systems for Chaotic Financial Markets.New York:Willey.
11. Drucker, H.,Burges, C. J. C.,Kaufman, L.,Smola, A.,Vapnik, V. N.(1997).Support vector regression machines.Advances in Neural Information Processing Systems,9,155.
12. Fama, E. F.(1970).Efficient capital market: a review of theory and empirical.Journal of Finance,25,383-417.
13. Hsu, C. W.,Lin, C. C.,Lin, C. J.(2003).,National Taiwan University.
14. Huang, W.,Nakamori, Y.,Wang, S. Y.(2005).Forecasting stock market movement direction with support vector machine.Computer and Operations Research,32,2513-2522.
15. Kim, K. I.,Jung, K.,Park, S. H.,Kim, H. J.(2002).Support vector machines for texture classification.IEEE Transactions on Pattern Analysis and Machine Intelligence,24,1542-1550.
16. Kim, K.,Han, I.(2000).Genetic algorithm approach to feature discretization in artificial neural networks for the prediction of stock price index.Expert system with application,19,125-132.
17. Koike, A.,Takagi, T.(2004).Prediction of protein-protein interaction sites using support vector machines.Protein Engineering Design & Selection,17,165-173.
18. Lee, T. S.,Chen, N. J.,Chiu, C. C.(2003).Forecasting the opening cash price index using grey forecasting and neural networks: evidence from the SGX-DT MSCI Taiwan Index Futures Contracts.Computational Intelligence in Economics and Finance,151-170.
19. Lee, T. S.,Chiu, C. C.(2002).Neural network forecasting of an opening cash price index.International Journal of Systems Science,33,229-237.
20. Leigh, W.,Hightower, R.,Modani, N.(2005).Forecasting the New York stock exchange composite index with past price and interest rate on condition of volume spike.Expert Systems with Applications,28,1-8.
21. Norinder, U.(2003).Support vector machine models in drug design: applications to transport processes and QSAR using simplex optimizations and variable selection.Neurocomputing,55,337-346.
22. Pai, P. F.,Lin, C. S.(2005).Using support vector machines in forecasting production values of machinery industry in Taiwan.International Journal of Advanced Manufacturing Technology,27,205-210.
23. Phua, P. K. H.,Ming, D.,Lin, W.(2001).Neural network with genetically evolved algorithms for stocks prediction.Asia-pacific Journal of Operational Research,18,103-107.
24. Rumelhart E.,Hinton, G. E.,Williams, R. J.(1986).Learning Internal Representations by Error Propagation in Parallel Distributed Processing.Cambridge, MA:MIT Press.
25. Shin, K. S.,Lee, T. S.,Kim, H. J.(2005).An application of support vector machines in bankruptcy prediction model.Expert Systems with Applications,28,127-135.
26. Tang, Z.,Fishwick, P. A.(1993).Feed forward neural nets as models for time series forecasting.ORSA Journal on Computing,5,374-385.
27. Tay, F. E. H.,Cao, L.(2001).Application of support vector machines in financial time series forecasting.Omega,29,309-317.
28. Thissen, U.,Van Brakel, R.,Weijer, A. P.,Melssen, W. J.,Buydens, L. M. C.(2003).Using support vector machines for time series prediction.Chemometrics and Intelligent Laboratory Systems,69,35-49.
29. Vapnik, V.,Golowich, S.,Smola, A. J.(1997).Support vector method for function approximation, regression estimation, and signal processing.Advances in Neural Information Processing Systems,9,281-287.
30. Vellido, V. N.,Lisboa, P. J. G.,Vaughan, J.(1999).Neural networks in business: a survey of applications (1992-1998).Expert Systems with Applications,17,51-70.
31. Yaser, S.,Atiya, A. F.(1996).Introduction to financial forecasting.Applied Intelligence,6,205-213.
32. 王彥茸(2000)。中央大學企業管理研究所。
33. 杜金龍(2006)。最新技術指標在台灣股市應用的訣竅。台北:財訊出版社。
34. 張政一(2000)。中國文化大學國際企業管理研究所碩士論文。中國文化大學國際企業管理研究所。
35. 莊智有(2000)。中原大學企業管理學系碩士論文。中原大學企業管理學系。
36. 陳國玄(2004)。成功大學統計研究所碩士論文。成功大學統計研究所。
37. 黃綺年(2004)。成功大學統計研究所碩士論文。成功大學統計研究所。