Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.411311
Title: Applied financial econometric analysis : the dynamics of swap spreads and the estimation of volatility
Author: Vives, David Mendez
Awarding Body: London School of Economics and Political Science
Current Institution: London School of Economics and Political Science (University of London)
Date of Award: 2003
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Please try the link below.
Access through Institution:
Abstract:
This Thesis contains an examination of the time-series properties of swap spreads, their relation with credit spreads and an estimation of the risk premium embedded in the swap spread curve. Chapter 2 introduces the main institutional aspects of swap markets, and studies the time-series properties of swap spreads. These are shown to be non-stationary and display a time-varying conditional volatility. Chapter 3 provides evidence of cointegration between corporate bond spreads and swap spreads. We estimate an error-correction model, including additional variables such as the level and slope of the yield curve, taking into account the exogenous structural break due to the crisis of August 1998. We find evidence that the relation between swap and credit spreads arises from the swap cash flows being indexed to Libor rates. Chapter 4 studies the risk premium in the term structure of the swap spreads, obtaining evidence that it is time-varying. The slope of the swap spread curve is shown to predict the changes in swap spreads. These results are relevant for the study of the risk premium in credit markets, and extend the existing literature on riskless Treasury securities. Chapter 6 develops the asymptotic properties of the quadratic variation estimator of the volatility of a continuous time diffusion process. We explore the case in which the number of observations tends to infinity, while the time between them remains fixed. For the case of a geometric Brownian motion, we show that the estimator is asymptotically biased, but the bias is a random variable that converges. We study the behaviour of this random variable via a simulation study, that shows that it typically has a "small" effect. We conclude by exploring some practical applications related the specification of the volatility for financial time series.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.411311  DOI: Not available
Share: