Why does skewness or the fat-tail effect play a major role in value-at-risk estimates?

What is it about?

This paper utilizes mainly the generalized autoregressive conditional heteroskedasticity (GARCH) model with the skewed generalized error distribution (SGED) to estimate the corresponding volatility and value-at-risk (VaR) measures for a range of commodities distributed across four types of commodities markets. The empirical results show that the return (volatility) of most of the assets distributed in alternative markets significantly decreased (increased) as a result of the global finance crisis. Conversely, owing to the oil crisis, there are not consistent results. As concerns the influential extent of both crises on return or volatility, the global financial crisis is more influential than the oil crisis. Moreover, with regard to all the confidence levels, the skewness effect actually exists in the VaR estimation only for the long position, whereas the fat-tail effect actually exists in the VaR estimation only for high confidence levels, irrespectively of whether a long or a short position was traded. Finally, with regard to the popular confidence levels in risk management, the SGED (GED) is the best return distribution setting for long (short) position. These results can provide the traders to precisely forecast the VaR and further to adjust appropriately the asset allocation they invested, and provide the authorities to make some policy to prevent this crisis to happen or lower the impact the crisis created.

Featured Image