On the Impact of Heavy-Tailed Returns to Popular Risk Measures: Evidence from Global Indices

Abstract

We use alpha-Stable Paretian assumption to estimate popular Drawdown-type and VaR-type risk measures by testing eight global equity indices for about ten years worth of data. The goodness-of-fit testing for the Stable model has been proved by the Kolmogorov-Smirnov methodology. We analyze the impact of several key factors such as kurtosis, skewness and tracking time on six risk measures: Maximum Drawndown, Conditional Drawdown at Risk, Drawdown at Risk, Average Drawdown, Value at Risk, and Conditional Value at Risk. Our results find some common trends, as well as individual characteristics for these particular risk measures.

Keywords: Heavy-Tails, Stable Paretian, Drawdown, Value at Risk, Expected Shortfall, Global Markets

JEL Classification: C16, C52, G15