The paper considers Edgeworth expansions for estimators of volatility.
Unlike the usual exapsions, we have found that in order to obtain
meaningful terms, one needs to let the size of the noise to go zero
asymptotically. This is reflected in our expansions. The results have
application to Cornish-Fisher inversion and bootstrapping.
KEY WORDS: Bootstrapping; Edgeworth expasion; Measurement error;
Subsampling; Market Microstructure;
Martingale; Bias-correction; Realized volatility