風險指標之實證研究-以台灣類股資料為例

Abstract

自2008金融風暴以降，風險管理成為重要的議題，風險管理其中重要的一環即為將風險量化以衡量風險的大小。本文以Robert J. Aumann與Roberto Serrano於2008年發表於Journal of Political Economy的文章中提出的新方法來衡量台灣各類股的風險，並同時計算台灣各類股的Value at Risk(VaR)與Conditional Value at Risk(CVaR)，比較新的風險指標與目前市場上常使用的VaR、CVaR在排序上有何不同，本文發現使用常態分配閉型解所求出的R值，與使用樣本資料原始分配反求出來的R值結果十分接近；在不同風險指標的排序上面，因新風險指標與VaR、CVaR要捕捉的風險不同，因此排序上也有所差異，因此在選擇衡量風險的方法時，應先考慮要捕捉的風險為何；最後，新風險指標必須在報酬率期望值大於零時才能計算，因此在衡量風暴期間各類股的風險上，較不具實用性。

Risk management has become important issue since 2008 financial tsunami. One of the most important parts of risk management is how to evaluate risk. This study adopts the method which was demonstrated by Robert J. Aumann and Roberto Serrano in 2008 to calculate the risk index (R index) of Taiwan sector stocks. Value at Risk (VaR) and Conditional Value at Risk of Taiwan sector stocks are calculated to be compared with the risk index. The conclusions are the followings: The result of R index under normal distribution assumption of Taiwan sector stocks is similar to the result of under the original distribution assumption. The rank of R index and the rank of VaR and CVaR are different because these methods evaluate different risks. Finally, the R index can not be used when the mean of the return distribution is negative. Therefore, when financial tsunami happens, the adoption of the R index would be difficult.