Comment.

This article comments on the article "Realized Variance and Market Microstructure Noise," by Peter R. Hansen and Asger Lunde. The authors make two arguments regarding Hansen and Lunde's article. They believe that corrections for microstructure noise do not matter in predicting future volatility and that the best predictor is power variation uncorrected for microstructure noise. Their comments are divided into three sections. The first section deals with decomposition issues, the second covers volatility forecasting and the third focuses on volatility measures other than increments in quadratic variation.