Value-at risk under normal assumption and extreme value theory

Abstract

In this paper, we wanted to compare and contrast the VaR estimates under the Normal assumption and the Extreme Value Theory for a selected portfolio of Singapore stocks. The methodologies chosen for the computations are the Variance-Covariance method and Generalised Extreme Value distribution approach. The results of our study over the sample period show us that the VaR estimates under the Extreme Value Theory assumption outperform that of the estimations computed under the Normal assumption.