標題: 信用投資組合法及信用循環指標法之比較研究
A Comparative Study of CreditPortfolio View and Credit Cycle Index Models
作者: 劉大安
Ta-Ann Liu
李正福
王克陸
Cheng-Few Lee
Keh-Luh Wang
經營管理研究所
關鍵字: 信用評等移轉矩陣;信用投資組合法;信用循環指標法;違約機率;Credit Migration Matrix;CreditPortfolio View;Credit Cycle Index;default probabilities
公開日期: 2005
摘要: 本研究主要的目的是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。在許多信用風險模型中,信用評等移轉矩陣,扮演關鍵的角色。本研究兼採Wilson(1997,信用投資組合法)以及Kim(1999,信用循環指標法)的研究方法,經適當調整,使用TCRI上市、上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,進而調整無條件信用評等移轉矩陣。實證結果發現,當實質GDP成長率上升時,信用循環指標法較能符合投機級違約率減少的趨勢,而當實質GDP成長率下降時,信用投資組合法較能符合投機級違約率上升的趨勢;此外,等級越低的違約率,波動度越高。回顧測試結果亦顯示,信用投資組合法較能吻合市場實際情況。
The purpose of this study is to investigate the impact on credit migration matrix by changes in business cycle, and calculate the related default probabilities. Credit migration matrix plays a crutial role in many credit risk models We modify Wilson (1997,CreditPortfolio View) and Kim (1999,Credit Cycle Index) 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 model and to adjust the unconditional credit migration probabilities. The major empirical results include: when business is in expansion, Credit Cycle Index model is more consistent with the expected decline in default probabilities for speculative classes; while during recession periods, PortfolioView model can provide higher default probabilities for risky firms In particular, lower rating firms exhibit higher volatility in default probability.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009337525
http://hdl.handle.net/11536/79655
Appears in Collections:Thesis


Files in This Item:

  1. 752501.pdf
  2. 752502.pdf
  3. 752503.pdf
  4. 752504.pdf
  5. 752505.pdf
  6. 752506.pdf
  7. 752507.pdf
  8. 752508.pdf
  9. 752509.pdf
  10. 752510.pdf
  11. 752511.pdf
  12. 752512.pdf
  13. 752513.pdf
  14. 752514.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.