臺灣博碩士論文加值系統

國人的理財觀念有逐漸普及且提升的趨勢，開始慢慢地體會到投資理財的重要性，大部分的投資理財書籍或相關網站都會給予投資人在進行投資前許多建議，但投資的策略以及相關的績效指標皆有相當多的選擇，對於不同的投資者而言，何種策略才是最好的投資方式以及如何選取其所對應之參考指標，並無一個明確的定論。
由於定期式的共同基金理財投資策略具有降低投資風險及平均投資成本的功效，因而受到基金投資人的青睞與採用，因此本研究針對國內共同基金市場進行分析與探討，使用民國96年至100年台灣所發行之所有種類的共同基金為樣本，並以較多投資人所使用之定期定額投資方式與定期不定額投資方式為基礎，進而設定不同報酬率下的贖回方式，找出各檔基金能獲得最佳報酬的投資方式進行分類，接著使用共同基金相關評估指標資料進行資料探勘，期望能找出不同投資策略下各相關績效指標的分類規則。
本研究除了以投資人的立場來探討共同基金之定期式投資策略分類問題外，更近一步地透過資料探勘技術中的決策樹方法探勘出在不同策略下，其相關績效指標之參考規則，研究結果發現，透過決策樹演算法分析後所得到的策略分類規則共有19條，希望能提供投資者在進行投資基金時當參考。

In society today, the idea of financial investment has become more prevalent as people begin to realize the importance of financial investment. The majority of the investment manuals or websites provide a lot of advice to prospective investors in making their investments. However, there are still many different strategies and relevant performance index that the investors could choose from. Unfortunately, for different types of investors, there is no single definite conclusion for what the best investment method is, and how to choose the corresponding index.
The monthly investment plan of mutual funds has the effect of lowering investment risks and balancing investment cost. It is especially favored by investors. Therefore, this study focuses on the analysis and discussion of the domestic mutual fund market. Every type of mutual fund issued in 2007 to 2010 in Taiwan is used as sample. In addition, the redemptions of different return on investment are based on dollar cost averaging and value averaging used by the majority of the investors. Furthermore, the investment approaches yielding the highest Return on Investment for every single fund are identified for the classification of strategies. Lastly, the research uses the assessment index of mutual fund to mining relevant information to understand the classification rules for the performance index of different strategies.
This study not only investigates the problems related to the systematic investment strategies in mutual fund from the investors’ view, it further employ decision tree based mining techniques to explore the classification rules of the relevant performance index under different strategies.
The research found that exploring information via the use of decision tree results in 19 classification rule of different strategies. It serves as a great reference to the investors.

誌謝 I
摘要 II
Abstract III
目錄 V
表目錄 VI
圖目錄 VII
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 11
1.3 研究範圍與對象 12
1.4 研究架構與流程 13
第二章 文獻回顧 15
2.1 投資組合問題之相關文獻 15
2.2 共同基金績效評估之相關文獻 18
2.3 資料探勘技術之相關文獻 22
第三章 研究方法 26
3.1 C5.0之決策樹演算法 26
3.2 Boosting技術 29
3.3 Pearson相關性分析 29
3.4 研究設計 30
第四章 實證結果與分析 40
4.1 測試例說明 40
4.2 共同基金之分類結果 42
4.3 測試例之探勘結果 43
第五章 結論與建議 50
5.1 研究結論 50
5.2 後續建議 51
參考文獻 52
附錄一 共同基金策略分類資料 55

一、英文部分
【1】A. Fernández and S. Gómez (2007) Portfolio selection using neural networks. Computers & Operations Research 34, 1177 – 1191.
【2】C. C. Lin and Y. T. Liu (2008) Genetic algorithms for portfolio selection problems with minimum transaction lots. European Journal of Operational Research 185, 393 – 404.
【3】G. Pye (1971) Minimax policies for selling an asset and dollar averaging. Management Science Vol. 17, No. 7, 379 – 393.
【4】H. Soleimani, H. R. Golmakani and M. H. Salimi (2007) Markowitz-based portfolio selection with minimum transaction lots, cardinality constraints and regarding sector capitalization using genetic algorithm. European Journal of Operational Research 180, 396 – 405.
【5】H. Markowitz (1952) Portfolio selection. The Journal of Finance, Vol.7, No.1, 77 – 91.
【6】I. Karagiannidis (2010) Management team structure and mutual fund performance. Int. Fin. Markets, Inst. and Money 20, 197 – 211.
【7】J. R. Quinlan (1993) C4.5：programs for machine learning. Morgan Kaufmann, San Mateo, CA.
【8】J. S. Chen, J. L. Hou, S. M. Wu and Y. W. C. Chien (2009) Constructing investment strategy portfolios by combination genetic algorithms. Expert Systems with Applications 36, 3824 – 3828.
【9】J. L. Treynor (1965) How to rate Management of investment funds. Harvard Business Review 13, 63 – 75.
【10】J. Han and M. Kamber (1999) Data Mining：Concepts and Technuques. Morgan Kaufmann Publisher.
【11】J. B. Cohen, E. D. Zinbarg and A. Zeikel (1986) Investment analysis and portfolio management. Irwin.
【12】K. J. Oh, T. Y. Kim, S. H. Min and H. Y. Lee (2006) Portfolio algorithm based on portfolio beta using genetic algorithm. Expert Systems with Applications 30, 527 – 534.
【13】K. Wang and S. Huang (2010) Using fast adaptive neural network classifier for mutual fund performance evaluation. Expert Systems with Applications 37, 6007 – 6011.
【14】L. H. Chen and L. Huang (2009) Portfolio optimization of equity mutual funds with fuzzy return rates and risks. Expert Systems with Applications 36, 3720 – 3727.
【15】M. M. Lai and S. H. Lau (2010) Evaluating mutual fund performance in an emerging Asian economy: The Malaysian experience. Journal of Asian Economics 21, 378 – 390.
【16】M. C. Jensen (1967) The performance of mutual funds in the period 1946 – 1964. Journal of Finance, Vol. 23, No. 2, 389 – 416.
【17】S. H. Liao, H. H. Ho and H. W. Lin (2008) Mining stock category association and cluster on Taiwan stock market. Expert Systems with Applications 35, 19 – 29.
【18】S. R. Nanda, B. Mahanty and M.K. Tiwari (2010) Clustering Indian stock market data for portfolio management. Expert Systems with Applications.
【19】S. P. Abeysekera and E. S. Rosenbloom (2000) A simulation model for deciding between Lump-sum and Dollar-cost Averaging. Journal of Financial Planning, Vol. 13, 86 – 96.
【20】T. Cura (2009) Particle swarm optimization approach to portfolio optimization. ENonlinear Analysis: Real World Applications 10, 2396 – 2406.
【21】T. J. Tsai, C. B. Yang and Y. H. Peng (2010) Genetic algorithms for the investment of the mutual fund with global trend indicator. Expert Systems with Applications.
【22】T. J. Chang, S. C. Yang and K. J. Chang (2009) Portfolio optimization problems in different risk measures using genetic algorithm. Expert Systems with Applications 36, 10529 – 10537.
【23】W. S. Lee, G. H. Tzeng, J. L. Guan, K. T. Chien and J. M. Huang (2009) Combined MCDM techniques for exploring stock selection based on Gordon model. Expert Systems with Applications 36, 6421 – 6430.
【24】W. F. Sharpe (1966) Mutual fund performance. The Journal of Business ,Vol. 39, No.1, 119 – 138.

【25】W. Chen and W. G. Zhang (2010) The admissible portfolio selection problem with transaction costs and an improved PSO algorithm. Physica A 389, 2070 – 2076.
【26】X. Huang (2008) Risk curve and fuzzy portfolio selection. Computers and Mathematics with Applications 55, 1102 – 1112.
【27】X. Li, Y. Zhang, H. S. Wong and Z. Qin (2009) A hybrid intelligent algorithm for portfolio selection problem with fuzzy returns. Journal of Computational and Applied Mathematics 233, 264 – 278.
【28】Y. W. C. Chien and Y. L. Chen (2010) Mining associative classification rules with stock trading data–A GA-based method. Knowledge-Based Systems 23, 605 – 614.
【29】Y. Chen, S. Mabu and K. Hirasawa (2010) A model of portfolio optimization using time adapting genetic network programming. Computers & Operations Research 37, 1697 – 1707.

二、中文部分
【30】邱顯比，基金理財的六堂課，初版，天下遠見出版股份有限公司，民國八十八年二月初版。
【31】牛田一雄、高井勉、木暮大輔原著，陳耀茂編審，資料採礦利用Clementine使用手冊，鼎茂圖書出版股份有限公司，民國九十五年二月初版。
【32】洪仁節、陳正佑、何怡滿、邱炳乾，理財規劃與理財工具，國立空中大學，民國九十七年十二月初版。
【33】薛薇、陳歡歌，Clementine 數據挖掘方法及應用，電子工業出版社，2010年9月初版。

三、網頁部分
【34】台灣經濟新報資料庫，http://www.tej.com.tw/twsite/
【35】中華民國證券投資信託暨顧問商業同業公會，
http://www.sitca.org.tw/
【36】行政院主計處，http://www.dgbas.gov.tw/