Optimization of credit portfolio by evolutionary algorithm

• Main Author: Yan, Ran
• Other Authors: Ponnuthurai Nagaratnam Suganthan
• Format: Final Year Project
• Language: English
• Published: 2009
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/17913
• Institution: Nanyang Technological University

During the past few decades, one of the most important advances in the investment field has been the creation of an optimum investment portfolio with desirable risk-return characteristics. The basic portfolio model was developed by Harry Markowitz, who derived the expected rate of return for a portfolio of assets and showed that the variance of the rate of return was a meaningful measure of portfolio risk. This portfolio variance formula not only indicated the importance of diversifying investments to reduce the total risk of a portfolio but also showed how to effectively diversify out the risks. Capital market theory extends portfolio theory and develops a model for pricing all risky assets. The final product, the Capital Asset Pricing Model (CAPM), will allow the investor to determine the required rate of return for any risky asset. To solve the credit optimization problem, evolutionary algorithms (EA) are always preferred. As EAs are able to find multiple Pareto-optimal solutions in one single run, development of evolutionary algorithms to solve multi-objective optimization problems has attracted much interest. Genetic algorithm (GA) is the most popular type of EAs. One seeks the solution of a problem in the form of strings of numbers, by applying operators such as selection, cross-over and mutation. GA is often used in optimization problems. In this report, Genetic Algorithm is used to construct optimal portfolio with desired combination of risk and return. Different risk models will also be studied and implemented in the portfolio optimization process. The constructed portfolio will be tested using back testing and stress testing for its performance. Additionally, GA’s control parameters will be studied to see how to improve GA’s performance and how long the investment strategy derived from the constructed portfolio can last.