本研究探討了收益率（earnings yield）和亞洲市場的長期政府債券收益率( long term government bond yield)的關係。並且運用結構性變異來以提高聯準會模型和股價的相關性。聯準會模型是用來判斷市場是否高估或低估股價或在其公允價值。本研究在亞洲十個主要市場進行實證研究，探討聯準會模型中不同的時間跨度的關係。結果顯示在亞洲國家，大盤的收益率和幾個月後的長期政府債券收益率之間有強關聯性。本研究通過迴歸分析研究來研究此模型的預測能力，並考慮不同的結構性變異檢定法ROC 和Bai_Perron檢定，結論顯示了ROC 檢定法更有效的偵測結構變異，提高聯準會模型的預測能力。 / This paper examines the possible relationship the earnings yield and long term government bond yield for the Asia markets. We apply structure break test to improve the Fed-model, which is used to judge whether stock prices are too high, too low or at their fair value. The paper examines the relationship proposed by the Fed- model with different time horizons. The findings reveal a strong association between long term government yield and the earnings yield in months later. The difference between the earnings yield and real bond yield is a shorthand measure for expected returns and we examine the predictive power of this measure by regression analysis. Considering ROC test and Bai_Perron test, it shows ROC test improves the forecasting power of Fed model with a better result.

本篇論文檢驗聯準會模型在六個亞洲市場：中國大陸、印度、馬來西亞、新加坡、台灣和泰國是否成立。我們首先檢驗共整合檢定來觀察變數之間長期的關係；另外，針對線性的指標模型，我們則檢測其是否具有非線性的門檻自回歸情形。實證結果顯示，於共整合檢定下，六個國家的股票價格、股票報酬和十年期債券殖利率具有長期共整合關係；而在非線性的TAR模型配適下，其解釋能力優於線性的AR模型。 / This paper studies the Fed Model in six Asia countries, China, India, Malaysia, Singapore, Taiwan, and Thailand. We examine the cointegraiton test for the long-run relationship and build a nonlinear threshold autoregressive model (TAR) between the long -term government bond yield, the stock index and the earning s index. Our empirical results show that such a long-run relationship indeed exists for those countries. In addition, the explanatory power of TAR model is better than linear AR model.