Content area
Full Text
(ProQuest Information and Learning: ... denotes formulae omitted.)
Over the last decade it has become more and more important to understand and model credit risk. Growing markets for corporate bonds and credit derivatives have stimulated development of advanced valuation and risk management models. The new Basel Capital Accord has also fueled the evolution of theoretical and applied research in credit risk.
Many of the new credit risk models require accurate observations of the term structure of interest rates of different credit risk classes as an input. Prominent examples are the Markov chain framework first introduced by Jarrow and Turnbull [1995] and extended byjarrow, Lando, and Turnbull [1997], and the class of reduced-form models that exogenously model the default intensity of a Poisson process as a function of stochastic state variables. This approach, introduced by Dume and Singleton [1997] and Lando [1998], enables the use of well-established results and techniques from the world of affine risk-free term structure models.
Both classes of models rely on the availability of reasonable data for term structures of credit spreads, i.e., the term structure of differences between risky term structures and the risk-free one. Apparently, there is a need for accurate and reliable procedures that estimate the term structure of credit spreads from observable coupon bond prices.
The estimation of the discount function or the zero-coupon yield curve from observable prices of coupon bonds is an established field of academic research. The most generally accepted procedures are due to pioneering work by McCulloch [1971, 1975], Schaefer [1981], Vasicek and Pong [1982], Nelson and Siegel [1987], and Langetieg and Smoot [1989]. More advanced estimation methods have been suggested by Steeley [1991], Svensson [1994|, Linton et al. [2001], and Subramanian [2001].
Interestingly, there are only rare contributions to the specific problems of credit spread estimation. Traditionally, credit spreads are calculated by subtracting independently estimated risk-free and risky term structures of interest rates, which in many cases yields unrealistically shaped and often irregular credit spread curves. Düllmann and Windfuhr [2004] and Geyer, Kossmeier, and Pichler [2004] report twisted credit spread curves for European Monetary Union government debt when the German curve is used as the risk-free reference curve.
Exhibit 1 shows a representative example where the yield curves for two government issuers...