Dependence Structure and Portfolio Value-At-Risk for Some Selected Nigerian Stocks Using a Copula-Based Volatility Model

Abstract

The study examined dependence structure and estimates portfolio risk on data from some selected Nigerian stocks. Marginal model for the stock returns and a joint model for the dependence for the dependence were specified. EVT model was employed for the marginal distribution of each return series, and for the joint model, the family of copula such as Gaussian, Frank, Gumbel, Clayton, BB7, Student-t copula were used with difference dependence structure, Using LL, AIC, and BIC values, BB7 is found to be the best fitted copula. Copula was used to measure Portfolio risk and global minimum risk portfolio is selected based on efficient frontiers. In estimating VaR, precise specification and identification of the probability of an extreme movement in the value of an individual asset (or portfolio) is essential for risk assessment. The evidence has direct implications for investors and risk managers during extreme currency market movements.

Keywords

References