Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization
Portfolio selection and optimization problems in the financial world have gained a lot of attention. The mean-variance model of the Markowitz (1959) has been widely applied to solve these problems, which considers the optimization process as an efficient diversification to obtain the optimal relatio...
| Main Author: | |
|---|---|
| Format: | Dissertation (University of Nottingham only) |
| Language: | English |
| Published: |
2009
|
| Online Access: | https://eprints.nottingham.ac.uk/23079/ |
| _version_ | 1848792505289539584 |
|---|---|
| author | ZHOU, LILI |
| author_facet | ZHOU, LILI |
| author_sort | ZHOU, LILI |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Portfolio selection and optimization problems in the financial world have gained a lot of attention. The mean-variance model of the Markowitz (1959) has been widely applied to solve these problems, which considers the optimization process as an efficient diversification to obtain the optimal relation between the expected return and risk (variance) of the structured portfolio. Portfolios of the lowest joint risk at a desired return or those of the maximal return at a certain risk are named efficient portfolios. Varying the desired portfolio return (or risk) provides opportunities to create the efficient frontier in a risk-return coordinate based on various efficient portfolios. Efficient frontiers give investors visible directions of what minimal risk matches their expected return or how great return they are able to reach at a level of risk, and therefore indicate the best way of investing money.
In this study, both Quadratic Programming (QP) and Genetic Algorithm (GA) methods are applied to find out the efficient portfolios based on which the efficient frontiers are traced out, considering the Markowitz mean-variance model. Efficient frontiers obtained from QP in EXCEL are used as the benchmark when testing the effectiveness of the proposed GA. Detailed discussions and valuable comparisons of the GA’s performances in various cases are undertaken. The robustness of the heuristic method is examined by real data sets collected from the FTSE 100 (UK) and the S&P 100 (USA). The empirical results show the GA can be successfully applied to obtain efficient portfolios without difficulty. It can provide a reliable result for portfolio optimization problems of different scales. |
| first_indexed | 2025-11-14T18:45:28Z |
| format | Dissertation (University of Nottingham only) |
| id | nottingham-23079 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T18:45:28Z |
| publishDate | 2009 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-230792018-02-15T14:26:00Z https://eprints.nottingham.ac.uk/23079/ Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization ZHOU, LILI Portfolio selection and optimization problems in the financial world have gained a lot of attention. The mean-variance model of the Markowitz (1959) has been widely applied to solve these problems, which considers the optimization process as an efficient diversification to obtain the optimal relation between the expected return and risk (variance) of the structured portfolio. Portfolios of the lowest joint risk at a desired return or those of the maximal return at a certain risk are named efficient portfolios. Varying the desired portfolio return (or risk) provides opportunities to create the efficient frontier in a risk-return coordinate based on various efficient portfolios. Efficient frontiers give investors visible directions of what minimal risk matches their expected return or how great return they are able to reach at a level of risk, and therefore indicate the best way of investing money. In this study, both Quadratic Programming (QP) and Genetic Algorithm (GA) methods are applied to find out the efficient portfolios based on which the efficient frontiers are traced out, considering the Markowitz mean-variance model. Efficient frontiers obtained from QP in EXCEL are used as the benchmark when testing the effectiveness of the proposed GA. Detailed discussions and valuable comparisons of the GA’s performances in various cases are undertaken. The robustness of the heuristic method is examined by real data sets collected from the FTSE 100 (UK) and the S&P 100 (USA). The empirical results show the GA can be successfully applied to obtain efficient portfolios without difficulty. It can provide a reliable result for portfolio optimization problems of different scales. 2009 Dissertation (University of Nottingham only) NonPeerReviewed application/pdf en https://eprints.nottingham.ac.uk/23079/1/Applicaitons_of_QP_and_GA_to_Portfolio_Optimizations.pdf ZHOU, LILI (2009) Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization. [Dissertation (University of Nottingham only)] (Unpublished) |
| spellingShingle | ZHOU, LILI Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization |
| title | Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization |
| title_full | Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization |
| title_fullStr | Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization |
| title_full_unstemmed | Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization |
| title_short | Applications of Quadratic Programming and Genetic Algorithm To Portfolio Optimization |
| title_sort | applications of quadratic programming and genetic algorithm to portfolio optimization |
| url | https://eprints.nottingham.ac.uk/23079/ |