Chaotic evolutionary multi-objective optimization for multivariate pair trading in tehran stock exchange: the distance approach

Pair formation is an important step in pair trading that has only been examined manually or through numerical instructions. These methods fail in the multivariate mode and do not consider conflicting goals in the problem structure. In this research, a method is presented to create multivariate pair combinations by considering contradictory multiple goals in stock pair trading.

In this study, the statistical sample is limited to the top 30 companies listed on the Tehran Stock Exchange due to the need for high-frequency transactions. The problem is developed in the form of a Mixed Integer Programming (MIP) model, and due to non-convex constraints and exponential solution space, a multi-objective genetic algorithm is used to obtain multivariate pair combinations. To achieve multiple goals, the developed type of genetic algorithm, namely, The Chaotic Non-dominated Sorting Genetic Algorithm (CNSGA-II), was used. In this method, chaos theory is used to create the initial population of the genetic algorithm in order to obtain appropriate and high-precision solutions.

The results showed that the use of chaos theory could increase the degree of convergence in evolutionary algorithms. In addition, these results indicate the superiority of the multi-objective pair trading strategy based on the distance approach over the traditional single-objective model.

Originality/Value: 

In order to optimize pair trading, the Non-dominated Sorting Genetic Algorithm was used. Also, the initial population of individuals was created in a multi-objective genetic algorithm based on chaos theory.