Edgeworth Corrections for Realized Volatility
研究了用埃奇沃思展开改进已实现波动率有限样本分布近似的方法，构建了积分波动率的置信区间，蒙特卡洛模拟显示修正区间优于正态近似区间，但自助法表现更好。
The quality of the asymptotic normality of realized volatility can be poor if sampling does not occur at very high frequencies. In this paper we consider an alternative approximation to the finite sample distribution of realized volatility based on Edgeworth expansions. In particular, we show how confidence intervals for integrated volatility can be constructed using these Edgeworth expansions. The Monte Carlo study we conduct shows that the intervals based on the Edgeworth corrections have improved properties relatively to the conventional intervals based on the normal approximation. Contrary to the bootstrap, the Edgeworth approach is an analytical approach that is easily implemented, without requiring any resampling of one’s data. A comparison between the bootstrap and the Edgeworth expansion shows that the bootstrap outperforms the Edgeworth corrected intervals. Thus, if we are willing to incur in the additional computational cost involved in computing bootstrap intervals, these are preferred over the Edgeworth intervals. Nevertheless, if we are not willing to incur in this additional cost, our results suggest that Edgeworth corrected intervals should replace the conventional intervals based on the first order normal approximation.