Use this URL to cite or link to this record in EThOS:
Title: Portfolio choice with independent components : applications in infrastructure investment
Author: Vermorken, M. A.
ISNI:       0000 0004 5363 8868
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
One of the principal questions in financial economics and applied finance relates to the optimal allocation of capital assets to portfolios. In recent times this field has received renewed attention as traditional portfolio optimisation methods were found to inadequately capture the nongaussian and interdependent nature of the returns of capital assets. A particular case is that of infrastructure assets, which exhibits particularly nongaussian and interdependent returns. In this thesis we introduce a portfolio choice method developed for nongaussian and interdependent assets and for longer investment horizons, as is common to infrastructure investment. Starting from the classical financial economic assumption of an expected utility maximizing investor, we derive an analytical solution, which incorporates all higher moments of the assets’ distributions without making limiting assumptions to ensure solvability. Rather than imposing subjective probability beliefs to infer the return’s distributions, we employ Independent Component Analysis to perform a decomposition of the asset space. In this way we are able to identify the fundamental drivers of the returns data and base our portfolio selection on their nature and interdependence. We apply the method on two samples of infrastructure assets. Firstly, we consider global infrastructure indexes. Secondly, we consider a large sample of airport operators, an asset class of particular interest to this thesis. In both cases we show how the method will outperform its principal rival and contestant, the standard mean-variance optimised portfolio. The thesis concludes by showing how the method also allows for a redefinition of the concept of diversification, fully integrated with the portfolio choice method. The thesis therefore contributes to the current state of the art and might lead to further research and discussion regarding the possible use of techniques like Independent Component Analysis to solve longstanding questions in theoretical and applied finance.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available