Stock Portfolio Optimization Using a Combined Approach of Relative Robust Risk Parity

Abstract

Risk parity is perceived as one of the stock portfolio selection models that have received a lot of attention since the US financial crisis in 2008. The philosophy of this model is to allocate the same amount of portfolio risk between the constituent assets. In the present study, the combined portfolio selection model of relative robust risk parity is introduced, which uses the worst-case scenario approach on the covariance matrix parameter appearing in the robust risk model in portfolio robustness. According to historical data, several scenarios are considered for the covariance matrix. The objective function value of the hybrid model for each portfolio (feasible point) is the worst result (with most volatility) among the set of scenarios. Finally, the model selects a portfolio for which the worst possible result has the least relative volatility. The research portfolio consists of 8 industries from Tehran Stock Exchange in the period 2011 to 2020. This portfolio has a higher Sharpe ratio than conventional models of mean-variance and weight parity, and is more resilient to market declines than the two models and produces less loss. Therefore, risk-averse investors are advised to use this stock portfolio selection model as a cover to face severe market declines.

Keywords

• risk parity portfolio
• relative robustness
• sharpe ratio
• particle swarm optimization algorithm