A bi-objective portfolio rebalancing model for index traking problem under transaction costs and solving it using meta-heuristic algorithms

Subject Areas : Financial Knowledge of Securities Analysis

Amir abass Najafi ¹ , Ehsan Fazeli Sabzevar ²

1 - Faculty of Industrial Engineering, K.N. Toosi University of Technology
2 - Faculty of Industrial Engineering, K.N. Toosi University of Technology

Received: 2014-02-01 Accepted : 2014-04-09 Published : 2014-12-22

Keywords: Index Tracking Portfolios Reba, portfolio optimization, Index tracking, Passive Management, Index fund,

Abstract :

Continuous rebalancing and optimization of the portfolio in a way that always leads to tracking the index accurately is a complex issue. Moreover, considering the transactional costs are inevitable. This paper proposes a model for optimization of the index tracking problem and a solution based on Genetic Algorithm. The proposed model is a bi-objective model that is aimed to minimize the tracking error and transactional costs. Due to complexity of the model, the Non-dominated Sorting Genetic Algorithm and the Non-Dominated ranked Genetic Algorithm are used and evaluated to solve the model. The basic metals Index of Tehran Stock Exchange at the year of 2012 is used in this research as the desired index for tracking and rebalancing.