Ranking Intraday Volatility Estimators Using Empirical Criteria

The problem of ranking intra-day volatility estimators purely via empirical criteria is considered. Following the original work on this topic undertaken by Patton (2011a), the proposed methodology draws from the literature on loss-based forecast evaluation, but accounts for the inevitable dependencies that occur when ranking estimators as opposed to forecasts. The rankings are proven to be consistent given a general near-epoch-dependence condition for the true volatility process, and the theory is supported by numerous simulations. A brief empirical section is also included