建構兩階段多目標之類免疫支援向量廻歸模式於股價預測

A Two-Stage Multi-Objective Artificial Immune System-Based Support Vector Regression Model for Stock Index Prediction

10.6338/JDA.201510_10(5).0001

邱志洲(Chih-Chou Chiu)；蔡易潤(Yi-Jun Tsai)；呂奇傑(Chi-Jie Lu)

股價預測 ； 人工類免疫系統 ； 支援向量迴歸 ； 多目標最佳化 ； Stock index prediction ； artificial immune system ； support vector regression ； multi-objective optimization

Journal of Data Analysis

10卷5期（2015 / 10 / 01）

1 - 30

繁體中文

由於眾多金融商品的價格波動都與股價指數漲跌有高度的關聯性，因此股價指數預測一直是財務時間序列預測領域最重要的議題之一。本研究基於人工類免疫系統（artificial immune system, AIS）與支援向量迴歸（support vector regression, SVR）提出一個兩階段多目標之類免疫支援向量迴歸模式（artificial Immune System-based support vector regression, AIS2VR）於股價預測。AIS是一個受免疫學啟發，並透過模仿免疫功能、原則和模式來解決最佳化問題的自我適應系統；SVR是一個以統計學習理論為基礎的預測技術。所提方法主要是藉著AIS來協助SVR快速地進行參數最佳化設置，同時基於AIS具有記憶區的特性，以兩階段多目標的方式搜尋SVR參數，使建構之股價預測模式能兼顧最小化數值預測誤差和最大化漲跌方向預測正確率，進而增加SVR於股價資料預測的精準度。本研究以台灣股市、香港股市與美國道瓊等三個股票市場的收盤價為實證標的，將所提方法與三個單階段預測模式及網格搜尋法進行比較。結果顯示，所提的AIS2VR兩階段多目標模式的預測誤差低於其他三個單階段的方法以及網格搜尋法，並且方向預測正確率明顯較高，代表所提方法可以確保在較低的數值預測誤差下，提升對股市漲跌方向預測的正確率。

Stock index is one of the most important research issues in financial time series forecasting since its variation affects the prices of financial products. In this study, a two stage multi-objective artificial immune system-based support vector regression model (called AIS2VR) is proposed for stock index prediction. Artificial immune system (AIS) is a computational problem solving method inspired by metaphors of the natural immune system immune system. Support vector regression (SVR) is a machine learning forecasting tool based on statistical learning theory and structural risk minimization principle. In the proposed model, the AIS algorithm based a two stage scheme and multi-objective principle is applied to optimize the parameters of the SVR prediction model since the parameters of SVR must be carefully selected in establishing an effective and efficient SVR model. In order to evaluate the performance of the proposed AIS2VR approach, the closing indexes of the Taiwan stock exchange capitalization weighted stock index (TAIEX), Dow Jones index (DJI) of U.S. and Heng Seng index (HSI) of Hong Kong are used as illustrative examples. Experimental results showed that the proposed AIS2VR stock index prediction model outperforms the three single stage AIS-based SVR forecasting models and the traditional grid search method. It is an efficient and effective alternative for stock index forecasting.

1. Atsalakis, G.S.,Valavanis, K.P.(2009).Forecasting stock market short-term trends using a neuro-fuzzy based methodology.Expert Systems with Applications,36,10696-10707.
2. Cao, L.J.,Tay, F.E.H.(2003).Support Vector Machine with Adaptive Parameters in Financial Time Series Forecasting.IEEE Transactions on Neural Networks,14,1506-1518.
3. Chen, K. Y.,Ho, C.-H.(2005).An improved support vector regression modeling for taiwan stock exchange market weighted index forecasting.Proceedings of the International Conference on Neural Networks and Brain,Beijing China:
4. Cherkassky, V.,Ma, Y.(2004).Practical selection of SVM parameters and noise estimation for SVM regression.Neural Networks,17,113-126.
5. Chong, Y.T.,Chen, C. H.(2010).Management and forecast of dynamic customer needs: An artificial immune and neural system approach.Advanced Engineering Informatics,24,96-106.
6. De Castro, L.N.,Von Zuben, F.J.(2002).Learning and optimization using the clonal selection principle.IEEE Transactions on Evolutionary Computation,6,239-251.
7. De Castro, L.N.,Von Zuben, F.J.(2000).The clonal selection algorithm with engineering applications.Proceedings of GECCO,Las Vagas, Nevada:
8. Drucker, H.,Burges, C.J.C.,Kaufman, L.,Smola, A.,Vapnik, V.N.(1997).Support vector regression machines.Advances in Neural Information Processing Systems,9,155-161.
9. Duan, K.,Keerthi, S.,Poo, A.(2003).Evaluation of simple performance measures for tuning SVM hyperparameters.Neurocomputing,51,41-59.
10. Hart, E.,Ross, P.(2002).Exploiting the analogy between immunology and sparse distributed memories: A system for clustering non-stationary data.Proceedings of the 1st International Conference on Artificial Immune Systems (ICARIS),Canterbury, UK:
11. Harta, E.,Timmis, J.(2008).Application areas of AIS: The past, the present and the future.Applied Soft Computing,8,191-201.
12. Hong, W. C.,Lai, C. Y.,Hung, W. M.,Dong, Y.(2010).Electric load forecasting by SVR with chaotic ant swarm optimization.Proceedings of the 2010 IEEE Conference on Cybernetics and Intelligent Systems,Singapore:
13. Hong, W. C.,Pai, P. F.(2006).Predicting engine reliability by support vector machines.International Journal of Advanced Manufacturing Technology,28,154-161.
14. Hou, L.K.,Yang, Q.X.(2009).Study on parameters selection of LSSVR based on grid-diamond search method.Proceedings of the International Conference on Machine Learning and Cybernetics,Baoding, China:
15. Hsu, C.-W.,Chang, C.-C.,Lin, C.-J.(2010).Technical ReportTechnical Report,National Taiwan University.
16. Ju, Y.F.,Wu, S.W.(2010).Village electrical load prediction by genetic algorithm and SVR.Proceedings of the 2010 3rd IEEE International Conference on Computer Science and Information Technology,Chengdu, China:
17. Lee, T.S.,Chen, N.J.(2002).Investigating the information content of non-cash-trading index futures using neural networks.Expert Systems with Applications,22,225-234.
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
19. Lixing, D.,Jinhu, L.,Xuemei, L.,Lanlan,L.(2010).Support vector regression and ant colony optimization for HVAC cooling load prediction.Proceedings of the 2010 International Symposium on Computer Communication Control and Automation(3CA),Tainan, Taiwan:
20. Luo, F.,Wu, J.,Yan, K.(2010).A novel nonlinear combination model based on support vector machine for stock market prediction.Proceedings of the 2010 8th World Congress on Intelligent Control and Automation,Jinan, China:
21. Malkiel, B.G.(2005).Reflections on the efficient market hypothesis: 30 years later.The Financial Review,40,1-9.
22. Mobinia, M.,Mobinib, Z.,Rabbani, M.(2011).An Artificial Immune Algorithm for the project scheduling problem under resource constraints.Applied Soft Computing,11,1975-1982.
23. Mohandes, M.A.,Halawani, T.O.,Rehman, S.,Hussain, A.A.(2004).Support vector machines for wind speed prediction.Support vector machines for wind speed prediction,29,939-947.
24. Myk, E.,Unold, O.(2010).Mining fuzzy rules using an Artificial Immune System with fuzzy partition learning.Applied Soft Computing,11,1965-1974.
25. Pai, P.F.,Lin, C.S.,Hong, W.C.,Chen, C.T.(2006).A hybrid support vector machine regression for exchange rate prediction.International Journal of Information and Management Sciences,17,19-32.
26. Roh, T.H.(2007).Forecasting the volatility of stock price index.Expert Systems with Applications,33,916-922.
27. Sakthivel, V.P.,Bhuvaneswari, R.,Subramanian,S.(2010).Artificial immune system for parameter estimation of induction motor.Expert Systems with Applications,37,6109-6115.
28. Timmermann, A.,Granger, C.W.J.(2004).Efficient market hypothesis and forecasting.International Journal of Forecasting,20,15-27.
29. Timmis, J.,Hone, A.,Stibor, T.,Clark, E.(2008).Theoretical advances in artificial immune systems.Theoretical Computer Science,403,11-32.
30. Vapnik, V.,Golowich, S.,Smola, A.(1997).Support vector method for function approximation, regression estimation and signal processing.Advances in Neural Information Processing Systems
31. Zhao, S.,Wang, L.(2010).Support vector regression based on particle swarm optimization for rainfall forecasting.2010 Third International Joint Conference on Computational Science and Optimization,Anhui, China:

1. 劉泰男、葉怡成(2016)。以實驗計畫法與迴歸分析建構多因子選股系統。Journal Of Data Analysis，11(1)，167-206。