Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Predicting Default – Moody’s, Merton, Logit – which is more accurate?

Low, Lihui Valerie LU and Petrova, Tsvetelina LU (2013) NEKN02 20131
Department of Economics
Abstract (Swedish)
This thesis aims to investigate if Moody’s is more accurate than logistic regression and the Merton model in predicting default by firms issuing corporate debt after the financial crisis. This is achieved by comparing the Expected Default Probabilities (EDPs) in Moody’s migration table published in their annual default study reports against the calculated EDPs from logistic regression and the Merton model. Our sample consists of both healthy and defaulted firms from 2009 to 2012 and uses company information 1 year prior to default. The results show that logistic regression seems to be more suitable than the Merton model and Moody’s for usage by financial institutions and individual investors who are deciding on lending to specific firms... (More)
This thesis aims to investigate if Moody’s is more accurate than logistic regression and the Merton model in predicting default by firms issuing corporate debt after the financial crisis. This is achieved by comparing the Expected Default Probabilities (EDPs) in Moody’s migration table published in their annual default study reports against the calculated EDPs from logistic regression and the Merton model. Our sample consists of both healthy and defaulted firms from 2009 to 2012 and uses company information 1 year prior to default. The results show that logistic regression seems to be more suitable than the Merton model and Moody’s for usage by financial institutions and individual investors who are deciding on lending to specific firms and purchasing corporate bonds, as they require more specific information on these particular firms’ default probabilities. (Less)
Please use this url to cite or link to this publication:
author
Low, Lihui Valerie LU and Petrova, Tsvetelina LU
supervisor
organization
course
NEKN02 20131
year
type
H1 - Master's Degree (One Year)
subject
keywords
Credit Risk, Moody’s, Expected Default Probabilities (EDPs), Merton model, Logistic Regression, Financial Crisis
language
English
id
3801254
date added to LUP
2013-06-12 14:40:43
date last changed
2013-06-12 14:40:43
@misc{3801254,
  abstract     = {{This thesis aims to investigate if Moody’s is more accurate than logistic regression and the Merton model in predicting default by firms issuing corporate debt after the financial crisis. This is achieved by comparing the Expected Default Probabilities (EDPs) in Moody’s migration table published in their annual default study reports against the calculated EDPs from logistic regression and the Merton model. Our sample consists of both healthy and defaulted firms from 2009 to 2012 and uses company information 1 year prior to default. The results show that logistic regression seems to be more suitable than the Merton model and Moody’s for usage by financial institutions and individual investors who are deciding on lending to specific firms and purchasing corporate bonds, as they require more specific information on these particular firms’ default probabilities.}},
  author       = {{Low, Lihui Valerie and Petrova, Tsvetelina}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Predicting Default – Moody’s, Merton, Logit – which is more accurate?}},
  year         = {{2013}},
}