本研究探討如何利用遺傳演算法(Genetic Algorithm)及適當的適應性函數以改進之前固 定比例投資組合保險策略(CPPI)的方法，利用從 1973 年至 2012 年台灣股價加權指數進行 訓練與測試不同適應性函數所演化出之最佳模型後所求得之複利報酬率(RG)、最大下跌幅 度(MDD)與左尾偏動差(𝐿𝑃𝑀1)進行相關比較，並採用 accuracy 與 precision 評量各適應性函 數所產生的結果是否擁有高度的正確率與精確度，以期望利用這些統計的方式確認何者在 訓練期與大盤比較之勝負於測試期亦有相同結果。結果顯示使用 Calmar ratio 而獲得之模 型不但比傳統型 CPPI 之投資策略來的優異外，並具有比其他適應性函數較佳之報酬率、
MDD 與𝐿𝑃𝑀1；並且採用 accuracy 和 precision 之統計方法評估亦顯示利用 Calmar ratio 能 夠獲得最佳的結果。同時我們利用 Calmar ratio 做為適應性函數演化之最佳模型(改良之 CPPI)與傳統之 CPPI (分別採用每月進出場以及隨機時間進出場)進行比較，經比較過後得 到改良之 CPPI 策略的確遠優於傳統 CPPI。我們經本研究後可得到利用遺傳演算法所求出 之改良 CPPI 投資組合策略的確能夠創造出比傳統之 CPPI 投資策略更高的報酬與更低的風 險。
本研究中我們已利用 7 種技術指標及 6 種績效指標，於未來可加入更多技術及績效指 標以進行更為全面性之研究。本文因考慮到大部份投資人無法承受較大風險情況，故採 用 CPPI 此種較為穩定之投資組合策略，在未來研究中我們計劃利用不同之投資組合策略 進行搭配，以利求得在風險容忍度各層級上適合運用何種適應性函數以進一步改良模型 之績效。
參考文獻
[1] Acar, E. and James, S., “Maximum loss and maximum drawdown in financial markets,” Forecasting Financial Markets Conference, London, 1997.
[2] Appel, G., The moving average convergence-divergence trading method: advanced version, Scientific Investment Systems, 1985.
[3] Bawa, V. S., “Optimal rules for ordering uncertain prospects,” Journal of Financial Economics 2(1), 1975, pp. 95-121.
[4] Becker, Y. L., Fei, P. and Lester, A. M., “Stock selection: An innovative application of genetic programming methodology,” Genetic Programming Theory and Practice IV,
Springer, 2006, pp. 315-334.
[5] Bertrand, P. and Prigent, J. L., “Leveraged ETF and CPPI type Strategy: Theory, Simulation and Empirical Study,” 2011.
[6] Black, F. and Jones, R. C., “Simplifying portfolio insurance,” The Journal of Portfolio Management 14(1), 1987, pp. 48-51.
[7] Black, F. and Perold, A., “Theory of constant proportion portfolio insurance,” Journal of Economic Dynamics and Control 16(3), 1992, pp. 403-426.
[8] Bookstaber, R. and Langsam, J. A., “Portfolio insurance trading rules,” Journal of Futures Markets 8(1), 1988, pp. 15-31.
[9] Chekhlov, A., Uryasev, S. and Zabarankin, M., “Drawdown measure in portfolio optimization,” International Journal of Theoretical and Applied Finance 8(1), 2005, pp.
13-58.
[10] Chen, J. S. and Liao, B. P., “Piecewise nonlinear goal-directed CPPI strategy,” Expert Systems with Applications, 33(4), 2007, pp. 857-869.
[11] Chen, J. S. and Lin, Y. T., “A partitioned portfolio insurance strategy by a relational genetic algorithm,” Expert Systems with Applications, 36(2), 2009, pp. 2727-2734.
[12] De Jong, K. A., Spears, W. M. and Gordon, D. F., “Using genetic algorithms for concept learning,” Springer US, 1994, pp. 5-32.
[13] Dichtl, H. and Drobetz, W., “On the Popularity of the CPPI Strategy: A Behavioral-Finance-Based Explanation and Design Recommendations,” The Journal of Wealth
Manage-ment, 13(2), 2010, pp. 41-54.
[14] Fishburn, P. C., “Mean-risk analysis with risk associated with below-target returns,” The American Economic Review 67(2), 1977, pp. 116-126.
[15] Goldberg, D. E. and Deb, K., “A comparative analysis of selection schemes used in genetic algorithms,” Foundation of Genetic Algorithms, 1991, pp. 69-93.
[16] Goodwin, T. H., “The information ratio,” Journal of Financial Analysts, 1998, pp. 34-43.
[17] Granville, J. E., A strategy of daily stock market timing for maximum profit, Prentice-Hall, 1960.
[18] Hamidi, B., Maillet, B. and Prigent, J. L., “A Risk Management Approach for Portfolio Insurance Strategies,” Proceedings of the 1st EIF International Financial Research Forum,
Economica, 2009.
[19] Holland, J. H., Adaptation in natural and artificial systems: An introductory analysis with applications to biology, control, and artificial intelligence, U Michigan Press, 1975.
[20] Huang, C.-F., Hsieh, T.-N., Chang, B. R. and Chang, C.-H., “A Comparative Study of Regression and Evolution-Based Stock Selection Models for Investor Sentiment,” Proc. of The 3rd International Conference on Innovations in Bio-inspired Computing and
Applications, Kaohsiung, 2012, pp. 73-78.
[21] Hunter, J. S., “The exponentially weighted moving average,” Journal of Quality Technology 18(4), 1986, pp. 203-210.
[22] Kaplan, P. D. and Knowles, J. A., “Kappa: A generalized downside risk-adjusted performance measure,” Journal of Performance Aeasurement. 8, 2004, pp. 42-54.
[23] Khuman, A. and Constantinou, N., How does CPPI perform against the simplest guarantee strategies, Working Paper, Centre for Computational Finance and Economic Agents
(CCFEA), 2009.
[24] Lane, G. C., “Lane’s stochastics,” Technical Analysis of Stocks and Commodities 2(3), 1984, p. 80.
[25] Lin, Y.-F., Huang, C.-F. and Tseng, V.-S., “A Novel Methodology for Stock Investment Using Episode Mining and Technical Indicators,” Proc. of 2012 Conference on Technologies
and Applications of Artificial Intelligence, Tainan, 2012, pp. 5-10.
[26] Linsmeier, T. J. and Pearson, N. D., “Value at risk,” Journal of Financial Analysts, 2000, pp.47-67.
[27] Magdon-Ismail, M., Atiya, A., Pratap, A. and Abu-Mostafa, Y., “On the maximum drawdown of a Brownian motion,” Journal of Applied Probability 41, 2004, pp. 147-161.
[28] Maringer, D. and Ramtohul, T., “GP-based rebalancing triggers for the CPPI,” Proc. of The 2011 Symposium on Computational Intelligence for Financial Engineering and Economics,
Paris, 2011, pp.1-8.
[29] Orito, Y. and Yamazaki, G., “Index fund portfolio selection by using GA,” Proc. of The 4th International Conference on Computational Intelligence and Multimedia Applications,
Yokusika City, 2001, pp. 118-122.
[30] Perold, F. and Sharpe, W. F., “Dynamic strategies for asset allocation,” Financial Analysts Journal, 1988, pp. 16-27.
[31] Pézier, J. and Scheller, J., “Best portfolio insurance for long-term investment strategies in realistic conditions,” Insurance: Mathematics and Economics 52(2), 2013, pp. 263-274.
[32] Prigent, J. L. and Tahar, F., “Optimal portfolios with guarantee at maturity: computation and comparison,” International Journal of Business 11(2), 2006, pp. 171-185.
[33] Balder, S., Brandl, M. and Mahayni, A., “Effectiveness of CPPI strategies under discrete-time trading,” Journal of Economic Dynamics and Control 33(1), 2009, pp. 204-220.
[34] Shadwick, W. F. and Keating, C., “A universal performance measure,” Journal of Performance Measurement 6(3), 2002, pp. 59–84.
[35] Sharpe, W. F., “Mutual fund performance,” Journal of Business 39(1), 1966, pp. 119-138.
[36] Sortino, F. A. and Van Der Meer, R., “Downside risk,” The Journal of Portfolio Management 17(4), 1991, pp. 27-31.
[37] Wong, W. K., Manzur, M. and Chew, B. K., “How rewarding is technical analysis? Evidence from Singapore stock market,” Applied Financial Economics 13(7), 2003, pp. 543-551.
[38] Young, T., “Calmar ratio: A smoother tool,” Futures Magazine 20(10), 1991, p. 40.
[39] Zhou, C., Yu, L., Huang, T., Wang, S. and Lai, K. K., “Selecting valuable stock using genetic algorithm,” Simulated Evolution and Learning. Springer Berlin Heidelberg, 2006, pp.
688-694.