Extreme value analysis and risk assessment: A case of pakistan stock market

Abstract

The aim of this paper consists in developing a univariate extreme value analysis, with applications to the Pakistan Stock Exchange (PSX-100). The main focus refers to assessing the risk and statistical properties regarding the tails of the fitted heavy-tails distributions. For this purpose, we implement generalized extreme value distribution (GEV) and generalized Pareto distribution (GPD) by following the block maxima approach, peak-over threshold (POT) method and Poisson processes to several declustered periods. PSX-100 has become the emerging and best performer financial market in the south-Asian region in the last decade. Therefore, statistical properties of extreme events of the stock market have significant importance for investors. We have also addressed the modelling of Value-at-Risk (VaR) and Expected-Shortfall (ES) risk measures, in the context of extreme value theory (EVT).