考量交易成本及允許放空之最差情況Omega比例模型

資產報酬的不確定性，可能誤導資產分配的預測，以至於投資組合再調整機制會形成效或低的績效。穩健型投資組合處理資產報酬的不確定性，其考量不同情境之報酬。近年來，研究者已經結合穩健型機制於條件風險值 (Conditional Value-at-Risk, CVaR) 的投資組合選擇，其中極端的損失被視為風險規避 (Zhu and Fukushima, 2009; Huang et al., 2010) 。然而，如果期望報酬大於期望損失則極端的損失可以被忽略，因Kapsos et al. (2014) 提出的最差情況Omega比例模型 (Worst-case robust Omega ratio, WCOmega) 作為CVaR的延伸，結合投資組合報酬的最大化和投資組合損失的最小化 (Kapsos et al., 2014)。本研究提出修改WCOmega投資組合選擇模型，並考量交易成本和放空來模擬金融市場的實際投資。本研究以最差情況Omega比例模型與最差情況條件風險值模型 (Worst-case Conditional Value-at-Risk, WCVaR)和相對穩健型條件風險值模型 (Relative Robust Conditional Value-at-Risk, RRCVaR)進行比較，透過夏普指數、Omega比例、權重、放空比例和交易成本，探討市場價值和實際報酬的表現。此外，這些模型之間的相似性，說明該模型之間有不同的資產配置特性。在不同市場趨勢的兩組資料，基金和標準普爾500指數 (S&P500) 的成份股用於顯示比較模型之間特性。結果顯示WCOmega模型之績效優於改良的CVaR模型。

關鍵字

穩健型投資組合；不確定性；最差情境Omega比例；最差情境條件風險值；相對穩健型條件風險值；再調整；放空；交易成本

並列摘要

The high uncertainty of asset returns could mislead the predictions of asset allocations so that the portfolio rebalance could be inefficiency or low performance. The robust portfolios deal with uncertainty based on the scenarios. Researchers have integrated the robust mechanism into Conditional Value-at-Risk (CVaR) for portfolio selection, which the extreme loss is considered for risk averison (Zhu and Fukushima, 2009; Huang et al., 2010). However, the extreme loss could be ignored if the expected return is higher than the expected loss. Because Worst-case robust Omega ratio model (WCOmega, Kapsos et al., 2014) extends the CVaR by integrating the maximization of portfolio returns and the minimization of portfolio loss (Kapsos et al., 2014). This study proposes the portfolio selection model which revises WCOmega model and considers transaction cost and short selling to simulate the real investment in the financial market. The comparsion of the proposed model with two other robust portfolio models the Worst-case Conditional Value-at-Risk (WCVaR) and the Relative Robust Conditional Value-at-Risk (RRCVaR) is based on Sharpe ratio, Omega ratio, weight, the ratio of short selling and transaction cost to explore the performance of market value and realized return. Moreover, the similarity among these models state the differences of asset allocation which is determined by the model characteristics. Two data sets, the funds and Standard & Poor's 500 index (S&P 500) are used to show the characteristics of the model comparison under different market trends. The results show that the proposed WCOmega has better performance than the modified CVaR related models.

並列關鍵字

Robust portfolio selections ； Uncertainty ； Worst-case Omega ratio ； Worst-case Conditional Value-at-Risk ； Relative Robust Conditional Value-at-Risk ； Rebalancing ； Short selling ； Transaction cost

參考文獻

Artzner, P., Delbaen, F., Eber, J. M., and Heath, D. (1999). Coherent measures of risk. Mathematical Finance, 9, 203–228.

Google Scholar

Bardong, F., Bartram, S. M., Yadav, P. K. (2008). Are short-sellers different? EFA 2008 Athens Meetings paper, Retrieved from http://ssrn.com/abstract=1101786.