Title: Two-stage portfolio risk optimisation based on MVO model
Authors: Krassimira Stoyanova; Vassil Guliashki
Addresses: Institute of Information and Communication Technologies – BAS, 'Acad. G. Bonchev' Str. Bl. 2, 1113 Sofia, Bulgaria ' Institute of Information and Communication Technologies – BAS, 'Acad. G. Bonchev' Str. Bl. 2, 1113 Sofia, Bulgaria
Abstract: This paper presents a two-stage portfolio risk optimisation based on Markowitz's mean variance optimisation (MVO) model. Historical return data for six asset classes are used to calculate the optimal proportions of assets, included in a portfolio, so that the expected return of each asset is no less than in advance given target value. Optimisation procedure is performed at the first stage, in order to select a limited number of assets among a large assets sample. At the second stage the optimal proportions of selected assets in the portfolio are calculated, minimising a risk objective function for a given rate of return. Ten optimisation problems are solved for different expected rate of return. The optimisation is performed in MATLAB. The proposed approach is robust and could be used successfully to solve large-scale portfolio optimisation problems.
Keywords: portfolio optimisation; mean variance optimisation model; MATLAB.
International Journal of Reasoning-based Intelligent Systems, 2020 Vol.12 No.1, pp.70 - 79
Received: 05 Nov 2018
Accepted: 20 Jun 2019
Published online: 10 Feb 2020 *