本研究使用基金績效、基金特性、基金經理人特質三種指標的組合，透過平滑支撐向量機建立基金選擇模型；並採用投資組合保險理論建立資金分配策略，再搭配遺傳演算法搜尋保險策略所需的參數，以建構股票式組合型基金。實證期間和對象為2002年10月至2007年9月間，台灣基金市場的147檔股票型基金。經實證研究發現：
(1)於選擇基金的階段加入基金經理人特質指標能提升平滑支撐向量基金選擇模型對基金的正確分類績效，在配合遺傳演算法搜尋投資組合保險策略所須使用的參數時，亦能得到較穩定的結果。
(2)所建立之組合型基金能有效且穩定提升累計報酬率，而在6次大盤指數報酬率為負的時期，前述的組合型基金之平均報酬率亦優於大盤指數，表示較能保護投資組合的價值。超額報酬的部份，則以於選擇基金的階段加入基金經理人特質指標，並搭配CPPI策略建立之組合型基金所能獲取的超額報酬最高，且變異係數最小。
(3)針對五年中大盤指數的各種盤勢進行比較，發現盤整期及空頭期間，前述搭配CPPI策略建立的組合型基金之平均年化月報酬率、Sharpe値的表現都優於大盤指數，而採行TIPP策略建立的組合型基金僅變異係數優於大盤指數。多頭期間，兩檔組合型基金的報酬率和Sharpe値都無法超越大盤指數。

We use smooth support vector machine (SSVM) to build fund-selection models. First, we combine fund performance, fund characteristics and fund managers’ characteristics to be input-variables. Second, we establish the portfolio insurance theory by using asset-allocation and genetic algorithm (GA), in order to build the fund of stock funds. Our empirical application is during 2002/10 to 2007/9, and the objects are 147 stock funds in Taiwan fund market. The conclusions are as following:
(1)Add the fund managers’ characteristics index in fund-selection stage, which is able to promote the classified performance of fund-selection model; it can obtain a more stable result when using GA to search the parameters.
(2)Fund of funds established by which mentions above can promote the accumulation return effectively and stably. When stock-index return rate are negative, the average return rate of fund of funds also superior to stock-index. It represented that our strategy can protect the value of the portfolio comparatively. In excess return, the fund of funds which add the fund managers’ characteristics index to the fund-selection stage and uses CPPI strategy, are able to gain highest excess return when the coefficient of variation is smallest.
(3)Finally, during market correction, the average year month return of fund of funds, coefficient of variation, and Sharpe are all superior to stock-index, but picking the line of TIPP strategy to establish the fund of funds is only coefficient of variation to superior to stock-index. In bull market, the return rate and Sharpe of two funds of funds are unable to surmount the stock-index. In bear market, the return rate, standard deviation and Sharpe of two funds of funds are all more outstanding.

中文部分：
[1] 中華民國證券投資信託暨顧問商業同業公會網站，http://www.sitca.org.tw
[2] 李純瑩（2004），「國內股票型基金績效之影響因素研究」，世新大學經濟學系碩士論文
[3] 林宜賢（2007），「應用平滑支撐向量迴歸與類神經網路於共同基金績效之預測」，國立台灣科技大學資訊管理研究所碩士論文
[4] 林政謙（2007），「應用平滑支撐向量機與類神經網路預測台灣上市電子指數漲跌之研究」，東吳大學經濟所碩士論文
[5] 林雅貞（2005），「組合型基金經理人投資行為之實證研究」，南台科技大學財務金融系碩士論文
[6] 吳聖修（2004），「應用股票趨勢技術分析於動態投資組合保險中之操作策略」，國立交通大學資訊管理研究所碩士論文
[7] 周萬順（2004），「共同基金在景氣循環下操作績效之研究」，世新大學經濟學系碩士論文
[8] 陳與泳（2004），「組合式基金資金分配策略－蟻元合作系統與遺傳演算法之應用」，國立台灣科技大學資訊管理研究所碩士論文
[9] 張正哲（2004），「共同基金選股能力與績效持續性之再驗證」，東海大學企業管理學系碩士論文
[10] 張嘉豪（2007），「應用平滑支撐向量分類於台灣股票市場選股之研究」，國立台灣科技大學資訊管理研究所碩士論文
[11] 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。
[12] 黃証國（2002），「應用基因演算法於動態投資組合保險中操作策略的最適化」，國立交通大學資訊管理研究所碩士論文
[13] 黃鈴惠（2006），「蟻元演算法應用於組合型基金商品設計之研究」，世新大學資訊管理研究所碩士論文
[14] 楊汝琬（2006），「績效超越大盤之基金其特性之研究」，國立中正大學財務金融研究所碩士論文
[15] 葉欣怡（2005），「共同基金經理人特質與持股特性之研究」，國立中正大學財務金融研究所碩士論文
[16] 劉育穎（2006），「結合決策樹與遺傳演算法建構不同風險程度之基金投資組合-以國內發行之股票基金為例」，中原大學資訊管理研究所碩士論文
[17] 劉貴強（2004），「遺傳演算法於組合型基金商品設計之研究」，輔仁大學資訊管理學系碩士論文
[18] 魏天元（2002），「考量總體資訊因素之基金擇時能力與基金特性之關係」，國立中山大學企業管理學系碩士論文
英文部分：
[1] Black, F. and Jonesm R. (1987), “Simplifying Portfolio Insurance”, The Journal of Portfolio Management, Vol.14 (1), pp.48-51.
[2] Black, F. and Perold, A. F. (1992),“Theory of Constant Proportion Portfolio Insurance”, Journal of Economic Dynamics and Control, Vol.16, pp.403-426.
[3] Bookstaber, R. and Langsam, J. (1988), ”Portfolio Insurance Trading Rules”, The Journal of Futures Markets, Vol.8, pp.15-31.
[4] Brennan, M. J. and .Schwartz E. S. (1988), ”Time-Invariant Portfolio Insurance Strategies”, Journal of Finance, Vol.43 (2), pp.283-299.
[5] Chen J.-S. and Chang C.-L. (2005), “Dynamical Proportion Portfolio Insurance with Genetic Programming”, Lecture Notes in Computer Science, Vol.3611, pp.735-743.
[6] Dahlquist, M., Engstroem, S. and Soederlind, P. (2000),. “Performance and Characteristics of Swedish Mutual Funds”, Journal of Financial and Quantitative Analysis, Vol.35 (3), pp.409-423.
[7] Estep, T. and Kritzman M. (1988), “TIPP: Insurance without Complexity”, Journal of Portfolio Management, Vol.14, pp.38-42.
[8] 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.
[9] Goldberg D.E. (1989), “Genetic Algorithms in Search, Optimization and Machine Learning”, Addison-Wesley Longman Publishing Co., Inc., Boston, MA.
[10] Golec, J. H. (1966), “The Effects of Mutual Fund Managers’ Characteristics on Their Portfolio Performance, Risk and Fees”, Financial Services Review, Vol.5, pp. 133-148.
[11] Henriksson, R. D. and R. C. Merton. 1981. "On Market Timing and Investment Performance II Statistical Procedures for Evaluating Forecasting Skill". Journal of Business, Vol.54, pp.513-533.
[12] Holland, J. H. (1975), “Adaption in Natural and Artificial Systems”, University of Michigan Press, Ann Arbor, MI.
[13] 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.
[14] Israelsen, C. L. (1998), “Characteristics of Winning Mutual Funds”, Journal of Financial Planning, Vol.11, pp.78-87.
[15] Jensen, M. C. (1968), “The Performance of Mutual Funds in the Period 1945-1964”, Journal of Finance, Vol.23, pp.385-416.
[16] 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.
[17] Sharpe, W. F. (1966), “Mutual Fund Performance”, Journal of Business, Vol.39, pp.119-138.
[18] Treynor,J. L.and Mazuy, K. K. (1966), “Can Mutual Fund Outguess the Market”, Harvard Business Review, Vol.44 , pp.131-136.
[19] Treynor, J. L. (1965), “How to Rate Management of Investment Funds”, Harvard Business Review, Vol.43, pp.63-75
[20] Wang T. and Rwegasira K. (2006), “Dynamic Securities Assets Allocation in Portfolio Insurance - the Application of CPPI and TIPP in the Chinese Capital Market”, Investment Management and Financial Innovations, Vol.3 (1), pp.97-103.
[21] Wermers, R. (2000), “Mutual Fund Performance: an Empirical Decomposition into Stock Picking Talent, Style, Transactions Costs and Expenses”, Journal of Finance, Vol.55, pp.1655-1695.