A Study on Portoflio Optimisation

HAN, YAN (2008) A Study on Portoflio Optimisation. [Dissertation (University of Nottingham only)] (Unpublished)

[img] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (471kB)


Portfolio optimization is a hot topic nowadays, this new type of investment research and analysis begin from the concept of Markowitz diversification theory. Portfolio theory examines the relationship that exists between risk and return when investing in equities. Obviously, investors are assumed to be risk averse, which means they wish to bear as little risk as possible for a given level of expected return, and the risk-reducing benefits make it wiser to invest in a diversified portfolio than in a single asset. Thereby, Markowitz theory which is also called Modern Portfolio Theory is based on the observation that where two assets are less than 100% positivity correlated, and thus the volatility of one will tend to cancel the volatility of the other and thus eliminate the non-systematic risk from an investor's portfolio. Nevertheless, many analysts has doubt the efficiency and reliability of Markowitz theory and then Genetic Algorithm is the one. In this study, Markowitz portfolio theory applications and GA portfolio approach has been both tested for this optimization problems. Accordingly, Markowitz portfolio theory has been examined in excel based on ten random selected securities with three different sectors, and GA portfolio approach model has been programmed in Maple 11 with the same ten securities and then twenty securities for further tested. The results indicated that Genetic Algorithm approach always got better results than Markowitz portfolio theory applications in terms of return divided by risks even there are some limitations existed. Thus the findings suggested that Genetic Algorithm could be a rational method for portfolio applications if research could further improve the performance of GA applications.

Item Type: Dissertation (University of Nottingham only)
Depositing User: EP, Services
Date Deposited: 11 Sep 2008
Last Modified: 24 Jan 2018 08:37
URI: https://eprints.nottingham.ac.uk/id/eprint/21860

Actions (Archive Staff Only)

Edit View Edit View