應用市場輪廓理論於CPPI與TIPP資產組合保險方法之整合分析

Abstract

本研究以市場輪廓理論為基礎，利用擺動因子企圖整合固定比例投資組合保險策略與時間無關投資組合保護策略，以臺灣50指數成分股為例，利用基因演算法最佳化固定比例投資組合保險策略與時間無關投資組合保護策略，結合擺動因子動態的調整每期的槓桿乘數、投資組合保險策略及風險部位的投資組合。 在基因演算法的適應函數部分提出報酬率與Sharpe Ratio兩種模式，為使有效選擇投資組合保險策略，擺動因子部分提出以擺動因子差值、擺動因子平均、擺動因子斜率，共六種模型，最後並比較單純固定比例投資組合保險策略與時間無關投資組合保護策略投資效益及準確率之差異性。 本研究提出應用市場輪廓理論之擺動因子解釋市場行為，嘗試進行主動式動態CPPI與TIPP投資組合保險策略之研究，經實驗分析與檢定後得出使用擺動因子動態模型優於只使用CPPI與TIPP之投資組合保險策略。保本底限較高時，當適應函數為報酬率時風險溢酬優於適應函數為Sharpe Ratio；在保本底限較低時，當適應函數為Sharpe Ratio時風險溢酬優於適應函數為報酬率。在曝險程度較高時，擺動因子皆能有效增加報酬；在曝險程度較低時，擺動因子差值跟平均的動態模型績效較佳。

The research is based on Market Profile, Conformation of CPPI and TIPP by using rotation factor, taking FTSE TWSE Taiwan 50 Index for example, it simulate CPPI and TIPP by using Genetic Algorithm, which combines the dynamic adjustment of rotation factor for Multiplier, Portfolio Insurance Policy and Portfolio of Exposure. It presents two models, which are Return and Sharpe ratio, in Fitness Function of Genetic Algorithm. For choosing Portfolio Insurance Policy, it figured out that unit as one year, average of six months and slope of six months, there are six models totally. Finally, it also compared the difference from efficiency and accuracy in CPPI and TIPP. It is proved by experiment, Conformation of CPPI and TIPP by using rotation factor better then CPPI and TIPP. In conclusion, at high floor, return is superior Sharpe Ratio in the risk premium; at lower floor, Sharpe Ratio is superior return in the risk premium. In a higher degree of exposure, the rotation factor increase the return effectively; at lower exposure levels, the rotation factor difference and average of the dynamic model performance is better.