Use this URL to cite or link to this record in EThOS:
Title: Modelling, forecasting and riding credit risk in the Sterling Eurobond market
Author: Manzoni, Katiuscia
ISNI:       0000 0001 3618 5869
Awarding Body: City University London
Current Institution: City, University of London
Date of Award: 2002
Availability of Full Text:
Access through EThOS:
Access through Institution:
This thesis aims to make a contribution to the understanding of credit risk dynamics in the Sterling Eurobond market. The background to the thesis is the increasing size, complexity and volatility of all debt markets, where the tasks of measuring, understanding and forecasting credit risk are of central importance to investing institutions and to corporate and sovereign borrowers. We investigate the changes in the perceived credit quality of bond issuers through three different approaches. First, we describe the evolution of credit spreads over time, exploring whether they reflect economic fundamental, or whether they represent self-generated force. This question is central to the fixed income literature in general, and to the pricing of risky debt and credit derivatives in particular. The time-series properties of our credit spreads provide strong evidence of mean-reversion, non-linearities, and directional and persistentv olatility. All these stylised facts are captured by time-varying volatility models. Second, we assess the information value of bond ratings, by examining the dynamics of bond spreads around rating revision dates. In contrast to standard event studies we apply a novel GARCH model to the panel data. This lets us examine the effects of the regrading event on the volatility of bond yields as well as the yields themselves. We find that downgrades are viewed as informational events, but upgrades are not. An asymmetric pattern is also observed in the dynamics of volatility. Third, we build a predictive structural model for the downgrade probability using a two-step estimation procedure. This allows us to disentangle the effects of credit rating and various financial and accounting ratios. We find evidence of non-linear effects from both company indebtedness and credit risk. The forecasting model is benchmarked against both a naive model, and a more sophisticated neural network model. Unlike the field of default prediction, little research has been done on forecasting the downgrade event. Filling this gap is of interest to banks and investors in periods of relative economic stability, in the context of value-at-risk models, and for the pricing of credit risky instruments.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: HG Finance Finance Taxation Management Mathematical statistics Operations research