Use this URL to cite or link to this record in EThOS:
Title: Essays on factor models : application to the energy markets
Author: Ipatova, Ekaterina
ISNI:       0000 0004 5365 5705
Awarding Body: City University London
Current Institution: City, University of London
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis focuses on the development of the theoretical, methodological and empirical literature on factor models. We provide detailed descriptions of the techniques used to estimate factor models, as well a means to establish the number of factors and assumption of factor models. The opening chapters address research from the theoretical investigation, which is motivated by the fact that for the past fifty years theoretical econometricians were working towards relaxation of the assumptions and increasing the consistency of the estimators. We offer an alternative solution which engineers faster rates of convergence for the estimated parameters, and furthermore without imposing any additional assumptions. The following chapter focusses on the problem of omitted observations in factor model datasets. Principle component analysis is only applicable to the balanced panel, therefore missing observations have to be filled. The modern literature predominantly focuses on the technique which can fill either missing observations at the beginning of the panel, or missing observations in the middle. Our technique offers a methodology which can help to fill missing observations irrespective of their place in the panel. Our technique is based on the factor model approach and uses factor model theory to develop the technique. The closing chapter focuses on empirical application of the factor models. We attempt to assess forecasting ability of the factor models in comparison with non-factor augmented counterparts and the univariate model. We use a robust approach which has never been applied to factor models and the crude oil market. Ultimately we show that the factor model approach can significantly improve forecasting ability in the crude oil market.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: HG Finance