信用風險評分模型區別力指標的預測研究

• Main Authors: He - Guei Chen, 陳和貴
• Other Authors: Yi-Ping Chang
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/51002106874233053442

碩士 === 東吳大學 === 商用數學系 === 95 === The New Basel Capital Accord (Basel II) permits banks to build their own credit rating model to reduce the risk capital. It's quite important whether the credit rating model is good or not. This research use the one factor model to simulate the obligors' score, the default model in Basel Committee on Banking Supervision to simulate the circumstances of default, and find the discriminatory power of the score model. Besides, this research set two different situations -- the first is the variables are predictable in the score model, and the second is not. The results find the first situation is better than the second situation. Besides, in both situations, the number of defaulted obligors would influence the stability of discriminatory power. The more defaulted obligors are more stable than the less. Increasing score correlation and asset correlation would increase the discriminatory power, but reduce the stability.