Estimation of the volatility parameter in value at risk (VaR) model

Abstract

In financial analysis, one of the most commonly used measures for evaluation of market risk is Value at Risk (VaR). Although it is an intuitively simple measure, estimating the underlying volatility can be quite complex. The main objective of the paper is to use Basic, EWMA and GARCH models for volatility parameter estimation in the Value at Risk (VaR) model. Using the real financial data, the methods are compared and determined which one gives the most appropriate estimate of the actual risk. The comparison of models is based on the analysis of the violation process. The results show that there is no single best method the best model depends on the data to be modelled.