Document Search
Add To My Bookshelf Sign in or Register Save And Annotate
Keywords:

credit sign in to follow this
creditmetrics sign in to follow this
portfolio sign in to follow this
factor sign in to follow this
non-public sign in to follow this
factor sign in to follow this
Merton sign in to follow this
structural sign in to follow this
Categories:

VaR Uses sign in to follow this
--Credit Risk sign in to follow this
Half-Life:
Impact:
Discuss This Paper
Sign in to follow this page
Recent Comments
  more
Calibrating the CreditMetrics Correlation Concept for Non-publicly Traded Corporations - Empirical E
Year Of Publication: 2003
Month Of Publication: July
Pages: 29
Download Count: 595
View Count:
Comment Num: 0
Language: EN
Source:
Who Can Read: Free
Date: 6-11-2006
Publisher: Administrator
Summary
This paper deals with the problem of empirically calibrating the CreditMetricsTM correlationconcept for a portfolio of loans to non-publicly-traded firms. Using this framework to determinethe probabilities of joint default events, requires an estimation of the portion of the obligors’ assetreturn volatility that is firm-specific (idiosyncratic). Whenever the bank’s obligors in thecredit portfolio under consideration are companies listed on a stock exchange, this can – in principle– be achieved by regressing individual stock returns on the returns of an appropriatelycomposed industry index. The resulting coefficient of determination (R-squared) for each companycan be interpreted as an estimate of the portion of return variation that is not firm-specific(systematic), i. e. caused by underlying factors which affect the firm’s industry as a whole. Butin the case of non-listed obligor firms, which is typical of many medium-sized enterprises inGermany (the so-called “Mittelstand”), this regression model cannot be fitted because of lackingstock price data. We analyze, how the solution to this problem currently offered in CreditManagerTM,which basically relates the weight of the idiosyncratic component to company size, canbe adapted to this case. We show that there is no empirically valid relationship between Rsquaredand market capitalization (book value of total assets) left, if those German stocks, whichare the heavy-weights in their CDAX? industry indices, are excluded from our random sample.Therefore, we suggest that a reasonable calibration of the CreditMetricsTM index model for Germannon-listed corporate obligors should do without a reference to compan
Author(s)
Hahnenstein, Lutz Sign in to follow this author
This document's citation network:
Similar Documents:
Close window
Sign up in one step, no personal information required. Already a Member?



Email:
Repeat Email:
User Name:
Password:
Confirm Password:

Sign Up


Welcome to GloriaMundi!
Thanks for singning up



continue or edit your profile