本論文以波動度偏離為基礎建立股票買賣的周策略,本文先驗證該策略的確能帶來超額報酬,再進一步提出使用產業分類方法來增進策略表現,最後計算不同策略條件在樣本時間的報酬表現。實作結果發現,波動度偏離策略本身的確具有一定預測力,根據波動度偏離分出的好壞分類兩群的報酬差距顯著異於零。如果使用減去產業平均的調整方法減去產業特性影響,可以使報酬提升,同時波動度顯著下降,而且此策略報酬並不能被Fama–French 三因子所解釋;若將此策略拉到產業層面,使用平均波動度偏離挑選產業,這時策略的報酬雖然會明顯提升,但卻伴隨著更高的波動度。
This thesis establishes a weekly strategy for stock trading based on volatility skewness. I first verify that the strategy can bring excess returns, and then further proposes to use industry classification to improve the performance. Finally, I calculate the performance of different strategic conditions at the sample time. The results of practice show that the volatility skewness strategy itself has a certain predictive power, and the difference between the rewards of the two extreme groups is significantly different from zero. If one uses the adjustment method, subtraction the industry average from individual skewness, to neutralize the influence of industry characteristics, one can increase the return and reduce the volatility significantly. In addition, the return of this strategy cannot be explained by the Fama – French three-factor model. However, it is not so effective to put this strategy on the industry level and use the industry-average volatility skewness to filter industries. Although returns from filtering industries will increase significantly, it is accompanied by much higher volatility.