情境變數改變可能會衝擊股市的報酬。本文所使用包含市場加總股利率、期間風險、倒帳風險、30天Gensaki債券附買回利率等情境變數衝擊的模型。結果發現該模型在日本股票市場解釋橫斷面報酬能力並沒有比Fama-French模型好。這些情境變數並不能替代Fama-French因子，因此HML和SMB仍保有它的解釋能力。日本市場自從1990年開始經濟衰退，因此本文亦進行子期間分析，試圖尋找1990年前和後之間的差別。結果顯示在1990年前SMB因子較顯著而HML在1990年後較顯著，但這兩個子期間仍沒有證據顯示情境變數的衝擊與Fama-French因子有關。 A model that includes shocks to aggregate dividend yield, term spread, default spread, and Gensaki rate do not explain the cross section of average returns better than the Fama-French model in Japanese market. The innovations to the four state variables do not proxy Fama-French factors, so HML and SMB do not lose their explanatory power for the cross-section returns. Since 1990, Japan have suffered economic recessions. Therefore, we conduct an subperiod analysis to investigate the difference between prior to 1990 and after 1990. It is interesting that SMB has an greater impact during January 1984 to December 1989 while HML has an greater influence during January 1990 to December 2003. However, like the overall period, these two subperiods do not offer much evidence that HML and SMB associate with the shock to state variables.
