Doctoral thesis

An empirical study of crude oil market


126 p

Thèse de doctorat: Università della Svizzera italiana, 2008 (jury note: magna cum laude)

English In this thesis I have tried to identify the risks and opportunities crude oil market offers. For this purpose I tested the performance of univariate and multivariate GARCH models. The first part of the work describes univariate GARCH models and their application to commodities markets. Physical ownership of the commodity carries an associated flow of services. The net flow of these services per unit of time is called 'convenience yield'. Since convenience yield appears as a factor which cannot be directly observed, the GARCH process was used for its modelling. The GARCH model has been found to provide a good fit for the convenience yield process. It paves the way for new term structure models of commodities prices and its empirical testing. The second part of my work uses multivariate GARCH models for hedging in oil market. For this reason, bivariate GARCH models have been estimated for cash and future prices. Taking into account the co-integration relation between these variables improves hedging performance. I test performance of co integrated GARCH model in and out of sample and find it superior to simple bivariate GARCH model.
  • English
License undefined
Persistent URL

Document views: 37 File downloads:
  • 2008ECO004.pdf: 8