The Dynamic Estimation of Default Probabilities for Taiwanese Unlisted Companies.

• Main Authors: Hua-Ching Kao, 高華慶
• Other Authors: 葉仕國
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/02988971140152082226

碩士 === 國立中興大學 === 財務金融系所 === 94 === With the implementation of New Basel Capital Accord, financial institutions have been setting up the internal rating-based model to measure the credit risk. One of the most important tasks is to estimate the default probabilities. Many researches of credit risk model applying accounting information cannot dynamically estimate a company’s default probability, while some other studies follow the popular Merton model limited by its difficulties and lack of the reliable historical default database, even though it can dynamically estimate the default probability that the estimation does not quantitate rationally. It can only for the listed companies but on the measurement of unlisted companies that there is not a dynamic and accurate estimation model at present. This article based on the Merton model tries to find the domestic corporations rated by the Standard & Poor’s which issues the rating default probabilities, estimating the listed companies’ default probabilities from the transformation of numerical method and combining the accounting information to construct the regression which can predict the default probabilities of unlisted companies. The variations of stock prices of listed companies can dynamically estimate the default probabilities of unlisted companies and the estimated default probabilities can approach actual default probabilities which meet the requirements under the New Basel Capital Accord.