如果報酬分配有系統性的偏斜(systematic skewness)，則預期報酬理當包含此偏斜風險所帶來的報償。 本論文主要目的為將系統共偏態(coskewness)加入多因子模型中，以驗證系統共偏態對資本資產定價是否有幫助。本研究以台灣證券市場上市公司的超額報酬為被解釋變數，擴充Fama and French(1993)和顧廣平(2005)所建構的因子模型，形成含有共偏態的多因子模式，並透過R2與GRS聯合檢定以檢驗共偏態效果於資產報酬定價方面是否有顯著的助益。 研究結果發現共偏態因子在多個因子模型中皆呈現顯著的狀態；且共偏態因子加入多因子模型後，對於超額報酬的共通時間序列變異解釋能力，有明顯的增加效果；而在解釋超額報酬的橫斷面變異方面，亦有不錯的表現。 If asset returns have systematic skewness, expected returns should include rewards for accepting this risk. The goal of this paper is to examine the linkage between the empirical evidence on these additional factors and systematic coskewness. Our study is to take the excess return of Taiwan Security Market as the dependent variable, and form some factor models by extending the factor models of Fama and French(1993) and the factor model of Kuang-Ping Ku(2005). We also test the co-skewness is helpful to capital asset pricing model by means of R2 and joint test of GRS. The empirical results indicate that conditional skewness has significantly effect on each factor model, and help to increase the explanation of the common time-series variation in returns. We also find that the co-skewness is helpful to the cross-sectional variation of expected returns.
