An empirical comparison of structural and accounting-based credit risk models.

• Main Authors: Leong, Wai Fun., Loh, Shueh Yi., Toh, Elaine Yi Lian.
• Other Authors: Leon Chuen Hwa
• Format: Final Year Project
• Language: English
• Published: 2009
• Subjects: DRNTU::Business::Finance::Risk management
• Online Access: http://hdl.handle.net/10356/15066
• Institution: Nanyang Technological University

This paper looks into 3 types of credit risk models: Altman Z-Score (2002), KMV-Merton (1974) and Longstaff and Schwartz (1997). These models are analyzed and tested using a sample of listed industrial firms in the Standard & Poor’s 500. With the aim of finding the most reliable and predictive model even in times of economic turbulence, a number of tests were carried out to test the correlation of the outputs calculated from the models against their market rating obtained from major Credit Rating Agencies. The qualitative and quantitative tests carried out in this report showed that KMV-Merton model has the strongest positive correlation with the corresponding Credit Rating Agencies’ ratings. A sensitivity analysis was carried out to test effect of a change in asset volatility on both KMV-Merton and Longstaff and Schwartz models. A simple prediction test examined the models on their ability to generate EDPs which can accurately predict the financial status of the firms for the next fiscal year.