Assessing the Accuracy of Default Prediction with Merton Model
Date Issued
2009
Date
2009
Author(s)
Jang, Chin
Abstract
We examine the accuracy of the Merton model, which is based on the theory proposed by Black and Scholes (1973) and Merton (1974). We compare the model with a simple alternative, which uses the same functional form with the Merton model but does not solve the model for an implied probability of default. We also compare the Merton style models with the model with the similar input but without the complex functional form and with the accounting models proposed by Altman (1968) and Lee (2002). We find that after considering the relevant market variables, the complex functional form of the Merton model and the numerical solution of the model do not contribute to the accuracy of default prediction. While contrary to the previous empirical result, the accuracy of the Merton model and other market models is definitely not inferior to the accuracy of either the accounting models. ur sample contains all Taiwan listed firms except for ones belong to the banking and finance industry. The time horizon is from 1990 to 2009. The full sample contains 1,689 Taiwan listed firms with 10,390 firm-year data. We divide the full sample into two subsamples according to the time. The main results hold in each subsample. We also conclude that the accuracy of both the accounting models and the market models improve as the time goes by.
Subjects
Credit Risk
Merton Model
Survival Analysis
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-98-R96723007-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):0692dc98dd0011e7e2674ea992a69a95