本研究探討景氣狀況對信用評等移轉機率的影響，進而計算其違約率。信用評等移轉矩陣，在許多信用風險模型中，扮演關鍵的角色。本研究兼Wilson (1997a, 1997b) 與Kim (1999) 的研究方法，經適當調整，使用TCRI 上市櫃公司信評等級以及1970-2004 年台灣總體經濟變數作為資料來源，以AR (1)-GARCH (1,1) 模擬總體模型，根據台灣之特有狀況選擇Seasonal ARMA 模型建構總體變數時間序列，進而透過總體經濟指標解釋違約機率，計算信用評等移轉矩陣之機率。實證結果發現，對投機級企業，當實質GDP 成長率上升時，信用循環指標法較能符合違約率減少的趨勢，而當實質GDP 成長率下降時，投資組合法較能符合違約率上升的趨勢；對投資型企業，二者都符合預期。此外，等級越低的違約率，波動度越高。

The purpose of this study is to investigate the impact on credit migration matrix due to changes in business cycle, and calculate related default probabilities. Credit migration matrix plays a crucial role in many credit risk models. We modify Wilson (1997a, 1997b) and Kim (1999), using TCRI credit rating data and various macroeconomic variables from 1970 to 2004, to estimate the transition matrix conditional on economic status. AR (1)-GARCH (1,1) model is proposed to simulate the macroeconomic variable and adjust the unconditional credit migration probabilities. The major empirical results reveal that, Credit Cycle Index model is more consistent with the expected decline in default probabilities for the speculative firms when business is in expansion; and Portfolio View model can provide higher default probabilities during recession periods. In particular, lower rating firms exhibit higher volatility in default probabilities.