Estimation of Value-at-Risk under jump dynamics and asymmetric information

This paper employs three Value-at-Risk (VaR) models (GARJI, ARJI and asymmetric GARCH) to compare the performance of 1-day-ahead VaR estimates. The influences of price jumps and asymmetric information on the performance of VaR are investigated. Two stock indices (Dow Jones and S&P 500) and one exchange rate (Japanese yen) are illustrated for estimating the model-based VaR. The results suggest for asset returns which exhibit time-variant jumps and information asymmetry, the VaR estimates generated by the GARJI and ARJI models provide reliable accuracy for low and high confidence levels. Moreover, as MRSB indicated, the GARJI model is more efficient than alternative models.