Realized Skewness at High Frequency and the Link to a Conditional Market Premium
32 Pages Posted: 25 Feb 2013 Last revised: 26 Apr 2014
Date Written: April 26, 2014
We investigate the asymptotic properties of an existing high frequency realized skewness measure and propose a more reliable new estimator which is robust to the microstructure noise at ultra-high frequency level. Asymptotic theory for the new estimator has been derived. Simulation example verifies the superior performance of the new estimator. We apply the new estimator with tick data of the S&P 500 index for forecasting one-month-ahead excess equity market returns in the U.S. from 1990-2011 and find robust and consistent result that realized skewness has significant forecastability both in-sample and out-of-sample. We also show that the new skewness measure plus the variance risk premium provides right decomposition for the skewness risk and it subsumes the market momentum effect in the short run. We thus provide valuable evidence that realized skewness has time series pricing ability in addition to cross sectional performance.
Keywords: Ito semi-martingale, High-frequency, Jump, Microstructure noise, Realized skewness, Stock return prediction
JEL Classification: C13, C14, G10, G12
Suggested Citation: Suggested Citation