A comparative study of range-based stock return volatility estimators for the German market

Abstract

This study investigates the relative performance of various volatility estimators based on daily and intraday price ranges of 25 German equities, with the two-scales realized volatility used as a benchmark. The empirical results show that all estimators based on daily ranges are by far superior to the classical estimator but are severely negatively biased due to discrete trading. The realized range obtained from intraday ranges performs better in terms of both bias and efficiency, although its performance still suffers from discrete trading. In these settings, the bias correcting procedure developed by Christensen and Podolskij (2007) appears to consistently outperform all other alternatives, including the scaled version of Martens and van Dijk (2007), and provides evidence of the relative advantages of the realized range