Abstract
Effective portfolio management requires a comprehensive understanding of the tradeoffs between different portfolio choices. Given a set of constraints, portfolio optimisation techniques based on linear programming and genetic algorithms can be applied to identify an optimal portfolio. However, this optimal portfolio might not be the preferred portfolio. Decision makers have to understand the tradeoffs between generally conflicting objectives and constraints before one portfolio can be identified as the preferred option. Such assessment of the overall search space is not made using linear programming and genetic algorithms when used to identify a single best solution and many portfolio options will have been eliminated before an understanding of these alternatives has been developed.
Markowitz's mean-variance approach and the traditional "rank and cut" approach are typically used to establish a relationship between a portfolio's value and its variance or associated development cost. Although these methods enable decision makers to compare and contrast different options, the optimisation is limited to the portfolio value measure and a single other metric. This latter limitation is overcome by the more recently developed portfolio filtering approach. This method is practical and transparent and allows for a quick development of strategic portfolio alternatives while considering a large number of portfolio attributes. Its main drawback is that the analysed set of portfolios generally represents a subset of the total search space. Hence, as the number of feasible portfolio options increases, so does the chance that the optimal portfolio is not present in the population of sampled portfolios.
This paper presents a portfolio optimisation approach that combines linear programming, genetic algorithms and portfolio filtering. An analysis of a realistic upstream portfolio illustrates and validates the optimisation process by demonstrating how limitations of individual methods can be mitigated by combining them.
