GJR-Copula-CVaR Model for Portfolio Optimization: Evidence for Emerging Stock Markets

This paper empirically examines the impact of dependence structure between the assets on the portfolio optimization, composed of Tehran Stock Exchange Price Index and Borsa Istanbul 100 Index. In this regard, the method of the Copula family functions is proposed as powerful and flexible tool to determine the structure of dependence. Finally, the impact of the dependence structure on the risk identification and the optimized portfolio selection, will be analyzed. The results show that the t-student copula function provides the best performance among other Copula functions. Also, empirical evidence suggests that the performance of the GJR-Copula-CVaR method is relatively more accurate and more flexible than other common methods of optimization.