Abstract
Algorithms based on the process of natural evolution are widely used to solve multi-objective optimization problems. In this paper we propose the agent-based co-evolutionary algorithm for multi-objective portfolio optimization. The proposed technique is compared experimentally to the genetic algorithm, co-evolutionary algorithm and a more classical approach—the trend-following algorithm. During the experiments historical data from the Warsaw Stock Exchange is used in order to assess the performance of the compared algorithms. Finally, we draw some conclusions from these experiments, showing the strong and weak points of all the techniques.
Highlights
The portfolio optimization problem is very important for every investor willing to risk their money in order to obtain potential benefits exceeding the average rate of profit of the capitalist economy.Before the 1950s, investors relied on common sense, experience or even premonitions in order to construct their portfolios
Algorithm 3: co-evolutionary multi-agent system (CoEMAS) pseudocode randomly INITIALIZE agents of two different sexes; foreach day do for round ← 1 to number_o f _rounds do foreach agent ∈ population do goal ← chooseGoal(); if goal is get resource perform actions hseek, geti leading to the realization of goal get resource end if goal is reproduce perform actions hseek, rep, rec, muti leading to the realization of goal reproduce end if goal is migrate perform action hmigi leading to the realization of goal migrate end end end end to Section 3.2, the mutation is used as a mean of maintaining the population diversity
In the second set of experiments we used the data from the year 2008—a year which was extremely hard for investors
Summary
The portfolio optimization problem is very important for every investor willing to risk their money in order to obtain potential benefits exceeding the average rate of profit of the capitalist economy. The model of co-evolutionary multi-agent system (CoEMAS), developed in our previous papers, allows for using many different biologically and socially inspired computation and simulation techniques and algorithms within one coherent agent-based system. The multi-objective portfolio optimization problem is used as a testbed for assessing the agent-based co-evolutionary multi-objective algorithm and the proposed technique for maintaining population diversity. This is only a small fragment of much broader research aiming at the formulation of a general model of agent-based systems for computing and simulation, utilizing biologically and socially inspired techniques and algorithms. The last part of the paper includes the results of two types of experiments, discussion of the results, and conclusions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
