[Back]

M. Frühwirth, P Schneider, L. Sögner:
"The Risk Microstructure of Corporate Bonds: A Bayesian Analysis";
Talk: 69. Wissenschaftliche Jahrestagung der Hochschullehrer für Betriebswirtschaft, Paderborn; 2007-05-31 - 2007-06-02.

This article presents joint econometric analysis of interest rate risk, issuer-specific
risk (credit risk) and bond-specific risk (liquidity risk) in a Lando (1998) type model
within the Duffie/Singleton framework. Given empirical bond data, the standard
procedure in the literature is to assume a benchmark bond with no bond specific risk
relative to the other bonds considered (see e.g. Duffie et al. (2003). We provide a
refinement of this approach by allowing for an identification of bond- and issuer
specific components for all bonds of an issuer. Since the number of bonds observed
is less than the number of latent processes an under-identification problem arises.
Our approach is to solve this problem by means of data augmentation, such that
exact Bayesian analysis can be performed. We apply our methodology to daily
default-risky bond data from the German market. The risk-free term structure is
inferred from liquid swap market data. Based on these estimates, issuer-specific and
bond-specific risk are estimated from corporate bond data. We observe that the bond
specific component differ significantly even for the same issuer. Last but not least,
we check for the determinants of the issuer and the bond specific components. We
observe that the risk-free term structure has an impact on the on the issuer-specific
and bond-specific spreads. Moreover, typical default proxies like the KMV distance to
default or the debt to value ratio also exhibit a weak impact on issuer and bondspecific
factors.