Value-at-Risk and market risk : overview and an application to the Athens stock exchange

Abstract

This paper provides an explanation of the significance of market risk for institutions and provides information on Value-at-Risk (VaR) methodologies as proposed by the Basel Committee on Banking Supervision and as implemented by institutions in order to measure this type of risk. It also analyzes the application of four VaR models, Historical Simulation with 100 observations, Historical Simulation with 250 observations and Variance-Covariance using Exponentially Weighted Moving Average and Simple Moving Average to forecast variance. These models are used to forecast the one-day ahead Value-at-Risk (VaR) of five one-asset portfolios based on data from the Athens Stock Exchange (ASE). By applying bactesting techniques as proposed by Christoffersen (1998) we confirm the basic weaknesses of these methods as stated in the related bibliography.