Use this URL to cite or link to this record in EThOS:
Title: Determinants of yield spreads on sovereign Eurobonds : an empirical study
Author: Audzeyeva, Alena
ISNI:       0000 0001 3432 778X
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
This thesis studies the detenninants of yield spreads on sovereign emerging Euromarket (EM) Eurobonds. It also contributes to the robust estimation of sovereign rating' transition matrices which are used as a measure of default risk in many financial applications. The empirical analysis here employs secondary market prices of individual Mexican, Colombian and Brazilian Eurobonds for the period 2003-2005. The analysis shows that default risk accounted for a rather small fraction (decreasing with maturity) of yield spreads for non-investment grade Colombian and Brazilian:· Eurobonds in 2003. This fraction increased while yield spreads fell significantly during 2003-2005 mainly due to non-default risk factors. The default and non-default risk components of investment-grade Mexican spreads both decreased at similar rates. Changes in the level (slope) of the U.S. interest rate curve are positively (negatively) related to movements in the zero-coupon EM yield spreads. The strength of the relationship between the U.S. short-term rate and yield spreads vary considerably for different EM countries. Brazilian Eurobonds are strongly affected by the fluctuations in the U.S. interest rate whereas Colombian Eurobonds show very little response. Global, region- and country-specific market risks are highly significant in explaining changes in EM yield spreads:· The model, incorporating these factors, explains up to 50% of variations in the E.M yield spreads during 2003-2005. An expected value estimator with 'expert knowledge' is introduced to the robust estimation of sovereign rat~g transitions which are characterised by few observations. The expected value estimates of default-components within Mexican, Colombian and Brazilian yield spreads provide a better fit than 'cohort' and continuous~time observation approaches. When the finite-sample properties of alternative estimators are compared, the contiriuous-time approach gives the most efficient estimates of default probabilities and default curves for investment-grade ratings. The expected value approach produces accurate estimates of qefault probabilities and default curves for some non-investment grade ratings.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available