Monte Carlo Valuation of Corporate Bonds Using Price Data

Corporate debt securities play a large part in financial markets and hence accurate modeling of the prices of these securities is integral. Ericsson and Reneby (2005) state that the corporate bond market in the US doubled between 1995 and 2005 and is now larger than the market for US treasuries. Although the theoretical corporate bond pricing literature is vast, very little empirical research to test the effectiveness of these models has been published. Corporate bond pricing models are split into two families of models. The first, are the structural models which endogenise default by modeling it as an event that may eventuate due to the insolvency of the underlying firm. The second family of models is the newer class of reduced-form models that exogenise default by modeling it as some random process (default intensity). The reduced-form models have been formulated largely due to the empirical failures of the structural family to accurately model prices and spreads. However as Ericsson and Reneby (2005) point out, an inadequate estimation approach may explain the poor performance of the structural models. Structural models are, therefore, the focus of this paper. We, however, do estimate a reduced-form model in order to make a comparison between the two types of model. There are no published papers (to my knowledge) in which both types of model are implemented ...