An Empirical Study: the Most Suitable Definition of Default Point of the KMV Model in Taiwan's Security Market

碩士 === 東吳大學 === 會計學系 === 94 === This study tries to define the most suitable default point (DP) of the expected default frequency (EDF) in Taiwan’s security market based on the Black and Scholes and Merton (BSM) contingent claims model, and Moody’s KMV Company framework. Since the estimates of the D...

Full description

Bibliographic Details
Main Authors: Yi-fen Liu, 劉怡芬
Other Authors: Da-bai Shen
Format: Others
Language:en_US
Online Access:http://ndltd.ncl.edu.tw/handle/08438566984068896986
Description
Summary:碩士 === 東吳大學 === 會計學系 === 94 === This study tries to define the most suitable default point (DP) of the expected default frequency (EDF) in Taiwan’s security market based on the Black and Scholes and Merton (BSM) contingent claims model, and Moody’s KMV Company framework. Since the estimates of the DP were seldom researched on this area, this study measures the parameters strictly. This study selects data from 1997 to 2004 to estimate the distance to default (DD) and the “risk neutral” default probabilities for a sample of TSE listed companies and GTSM (OTC) listed companies over period 1998 to 2005. This study chooses current liabilities, current liabilities plus a half of long-term liabilities, current liabilities plus long-term liabilities and net liabilities to substitute for traditional DP-total liabilities. This study also uses solvency ratio (SR) to substitute for asset market value and DP. The empirical results treat net liabilities and total liabilities as the best DP in the security market. However, when this two kind DP do not perform well, SR has better performances under the discussion of electronics and construction industries. The empirical results also find different kind of industries has different suitable DP, so does different kind of financial distresses. A type of DP can not suit all kinds of situations. When an institution builds a credit risk model, it should consider who the object it is, what purpose it is and so on. Therefore, it can establish a high quality model to predict the credit risk.