離散報酬與股價指數報酬關係之探討

本研究針對離散報酬是否可以作為資產定價模型的一個有力因子，進行深入的探討，並以台灣50指數、中型100指數、金融保險類指數、電子類指數為分析對象，選取2000年至2014年上述指數成份股報酬率，計算離散報酬率，再將離散報酬因子加入CAPM與Fama and French三因子模型，以及GARCH模型中，檢視其對模型的影響力與解釋能力。研究結果發現，離散報酬作為資產定價模型的因子，對於股價指數報酬確實具有顯著的影響及解釋能力，此外，實證顯示離散報酬(RD)對於股價指數未來走勢具有預測性，且以Fama and French(1993)所建構的六個投資組合所進行的迴歸分析發現，在加入離散報酬因子之後，模型可以提供更佳的解釋能力，表示在加入了離散報酬因子(RD)之後，可以使我們更準確的預測股價指數的報酬，總體而言，離散報酬因子(RD)在資產定價的研究上，確實具有代表性的地位。

關鍵字

離散報酬；資本資產定價模型；三因子模型； GARCH模型

並列摘要

This study explores whether the return dispersion could as a powerful factor in asset pricing model, and aim at index component stocks of return, including Taiwan 50 Index, medium 100 index, financial and insurance index, electronic index in 2000-2014 years as an sample. Then calculate return dispersion and add it into CAPM and Fama and French three-factor model and GARCH model, to examine its influence and explain ability to the models. The results found that return dispersion as factor in asset pricing model, indeed have significant influence and explain ability, besides, the results of study express return dispersion also can forecast future trend for stock index. Consulting Fama and French(1993) to constructing six portfolio, and we find while regression analyzing, that the model can offer extra explanation ability to the regression model after add to value at return dispersion factor. Express after add to the value at return dispersion factor, the expected stock index return enabling us to be more accurate. To the end, this means the return dispersion factor at the study of asset pricing have representative status.