USING «FAT TAILS» DISTRIBUTIONS FOR RISK MANAGEMENT

Abstract

Value at Risk (VaR) is one of the most important indicators used in risk management. Typically the normal distribution is used for its calculation, but the actual distribution of securities’ returns has much more probability in its tails than the normal distribution. In this article we derive VaR formulas for the Student’s t and Laplace distributions. Using assets on the U.S. and Russian stock market, we show that using these «fat-tailed» distributions lead to a significant increase in quality of risk measurement comparing to the normal distribution.