許多人工智慧方法常有過多參數需要調整與測試，使得模型複雜度增加，而平滑支撐向量機(Smooth support vector machine, SSVM)中，採用RBF核心函數僅需調整 和 兩個參數，相對於其它方法，具有較低模型複雜度。
過去研究運用相關預測工具於選股議題上，未考量實際取得財務比率之時間點，本研究考量實際上資料取得時點，依據其資料結合平滑支撐向量分類(Smooth support vector classification , SSVC)建立選股模型， 和 參數搜尋方法則分別採用Grid和UD，並擬定單期和多期參數搜尋策略，實證1996/6/31~2006/10/31期間，SSVC選股模型之績效是否具有超額報酬。
實證結果顯示，多期參數搜尋策略結合Grid參數搜尋法所建立之SSVC選股模型，於空頭或多頭市場，皆可獲得超額報酬。

Many methods of artificial intelligence often need to adjust too many parameters, tests and complexity. Smooth Support Vector Machine (SSVM), adapt the RBF adjustment and two parameters only, as opposed to other methods, SSVM with lower complexity.
Previous studies related predictive tools used in selecting stocks that, do not take into account the financial ratios achieved time. This study considers the timing of obtained data in reality, based on its data combination with Smoothing Support Vector Classification (SSVC), and establish stock-selection model. The methods of searching parameters use UD and Grid to develop the parameters searching strategy of single-phase and multi-phase. The empirical time is 1996/6/31~2006/10/31. The study discusses whether the performance of SSVC-Selection Model superior to excessive compensation or not.
The empirical results show that the SSVC-Selection Model with multi-phase strategy combining Grid, both in bear or bull markets can achieve excess return.

中文部份：
[1]吳智鴻(2004)，「結合基因演算法最佳化「支持向量機」參數∼財務危機上之應用」，國立台北大學企業管理研究所博士論文。
[2]徐美珍(2003)，「企業財務危機之預測」，國立政治大學統計研究所碩士論文。
[3]黃奎傑(2006)，「應用SSVM與灰色預測於投資策略之研究-以台灣股票市場電子類股為實證」，國立台灣科技大學資訊管理研究所碩士論文。
[4]黃建銘(2005)，「支撐向量機的自動模型選擇」，國立台灣科技大學資訊工程研究所碩士論文。

英文部份：
[1]Albanis, G. A. and Batchelor, R. A. (2001), “21 Nonlinear Ways to Beat the Stock Market”, Developments in Forecast Combination and Portfolio Choice, Available at http://www.staff.city.ac.uk/r.a.batchelor/21method.pdf .
[2]Ball, R. and Brown, P. (1968), “An Empirical Evaluation of Accounting Income Numbers”, Journal of Accounting Research, pp.663-681.
[3]Fama, E. F. (1970), “Efficient Capital Markets: A Review of Theory and Empirical Work”, Journal of Finance, Vol.25, pp.383-417.
[4]Fama, E. F. (1999), “Market Efficiency, Long-term Returns, and Behavioral Finance”, Journal of Financial Economics,Vol.49 (3), pp.283-306.
[5]Fan, A. and Palaniswami, M. (2001), “Stock Selection Using Support Vector Machines”, Proceedings of the International Joint Conference on Neural Networks, Vol.3, pp.1793-1798.
[6]Hsu, C.-W., Chang, C.-C. and Lin, C.-J. (2003), “A Practical Guide to Support Vector Classification . Technical Report”, Department of Computer Science and Information Engineering, National Taiwan University. Available at http://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf.
[7]Lee, Y.-J. and Mangasarian, O. L. (2001), “SSVM: A Smooth Support Vector Machine for Classification”, Computational Optimization and Applications, Vol.20, pp.5-22.
[8]Lee, Y.-J. and Huang, S.-Y. (2007), “Reduced Support Vector Machines: A Statistical Theory”, IEEE Transactions on Neural Networks, Vol.18 (Part 1), pp.1-13.
[9]Markowitz, H. M. (1952), “Portfolio Selection”, Journal of Finance, pp.77-91.
[10]Vapnik, V. N. (1995), “The Nature of Statistical Learning Theory”, New York: Springer-Verlag, pp.188.
[11]Vapnik, V. N. and Chervonenkis, A. Ya. (1974), “Ordered Risk Minimization – 1”, Automation and Remote Control, Vol.35 (8 Part 1), pp.1226-1235.
[12]Vapnik, V. N. and Chervonenkis, A. Ya. (1974), “Ordered Risk Minimization – 2”, Automation and Remote Control, Vol.35 (9 Part 1), pp.1403-1412.
[13]Veropoulos, K., Campbell, C. and Cristianini, N. (1999), “Controlling the Sensitivity of Support Vector Machine”, In Proceedings of the International Joint Conference on Artificial Intelligence.