Title: Portfolio optimisation with cardinality constraint based on expected shortfall
Authors: Ezra Putranda Setiawan; Dedi Rosadi
Addresses: Universitas Negeri Yogyakarta (UNY), Yogyakarta, Indonesia ' Mathematics Department, Universitas Gadjah Mada, Yogyakarta, Indonesia
Abstract: Assets diversification is a well-known strategy to reduce the investment risk and become a mathematical problem since Markowitz's work in 1952. In this paper, we investigated the portfolio selection method using expected shortfall (ES), which also known as expected tail loss (ETL) or conditional value-at-risk (CVaR), as a risk measure. A cardinality constraint was added to the model in order to help the investor choose k from n available assets into the portfolio, where k is higher than the lower bound L and smaller than the upper bound U. To solve this complex portfolio optimisation problem, we use the genetic algorithm method with binary chromosomes and obtain the optimal weight using exact method. A numerical case-study is provided using several stocks in Indonesia Stock Market.
Keywords: genetic algorithm; portfolio; cardinality constraint.
DOI: 10.1504/IJCSM.2020.111707
International Journal of Computing Science and Mathematics, 2020 Vol.12 No.3, pp.262 - 273
Received: 05 Jun 2018
Accepted: 22 Oct 2018
Published online: 11 Dec 2020 *