Journal of Statistics Applications & Probability

Author Country (or Countries)

South Africa


In the past decade, analysts, statisticians and researchers have become more interested in the research and applications of extreme value mixture models in the stock market and insurance as well as medical industries. This study aims to evaluate the fit of two extreme value mixture models namely GPD-Normal-GPD (GNG) and GPD-KDE-GPD (GKG), where KDE represents the Kernel density estimator, for three FTSE/JSE indices namely All Share Index (ALSI), Banks Index and Mining Index and the USD/ZAR currency exchange rate. Value at Risk (VaR) assesses market risk and many financial corporations often seek reliable VaR estimates. VaR estimates and the Kupiec likelihood backtesting procedure are calculated to evaluate the tail behaviour of the fitted GNG models. Results highlight the robustness of the GNG and GKG mixed model for each daily returns when compared to the traditional Normal model that is commonly applied model in financial literature. Financial practitioners looking to curb losses and explore alternatives for financial modeling in the South African financial industry using an extreme value mixed model approach may gain the most by implementing the GNG or GKG model

Digital Object Identifier (DOI)