Value-at-Risk Forecasts in U.S. Stock Market index with Skewed Generalized Error Distributions

What is it about?

This paper proposes a composite Simpson’s rule, which is a numerical integral method, for the estimation of VaR thresholds based on the skewed generalized error distribution(SGED). We examine two Value-at-risk(VaR) models(GARCH-N and GARCH-SGED), which have the normal and SGED innovations, to compare the performance of 1-day-ahead VaR estimates. Two stock indices(Dow Jones and S&P500) are illustrated for estimating the model-based VaR. The results suggest for asset returns which exhibit skewed to the left and leptokurtic, the VaR estimates generated by the GARCH-SGED model provide reliable accuracy for low and high confidence levels.

Featured Image