In this paper, by applying the potential approach to characterizing default risk, a class of simple affine and quadratic models is presented to provide a unifying framework of valuing both risk-free and defaultable bonds. It has been shown that the established models can accommodate the existing intensity based credit risk models, while incorporating a security-specific credit information factor to capture the idiosyncratic default risk as well as the one from market-wide influence. The models have been calibrated using the integrated data of both treasury rates and the average bond yields in different rating classes. Filtering technique and the quasi maximum likelihood estimator (QMLE) are applied jointly to the problem of estimating the structural parameters of the affine and quadratic models. The asymptotic properties of the QMLE are analyzed under two criteria: asymptotic optimality under the Kullback-Leibler criterion, and consistency. Relative empirical performance of the two models has been investigated. It turns out that the quadratic model outperforms the affine model in explaining the historical yield behavior of both Treasury and corporate bonds, while producing a larger error in fitting cross-sectional bond spread curves. Moreover, a modified fat-tail affine model is also proposed to improve the cross-sectional term structure fitting abilities of the existing models. Meanwhile, our empirical study provides complete estimates of risk-premia for both market risk and credit default risk including jump event risk.
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
page. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
Publisher Info
Paper provided by EconWPA in its series Finance with number
0312009.
References listed on IDEAS Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.: