Risk analysis of high frequency precious metals returns by using long memory model
This paper examines the long memory feature in the conditional volatility and Value at Risk calculations of precious metals returns at different time frequencies. In doing so, we employ high frequency data while using FIGARCH models. Furthermore, significant long memory characteristics have been detected in absolute returns by using semiparametric Local Whittle (LW) approximation. The empirical evidence of the local memory parameter across different time intervals gives consistent approximates of the long memory parameter which indicates that the series show some degree of self-similarity. Further, empirical results show that the long memory characteristics remains fully consistent across different time intervals of data for conditional and unconditional volatility measures. This study is useful for investors and traders (with different trading horizons) with regard to forecasting volatility and calculating or implementing trading strategies at different time frequencies.