The Realized Volatilities Research on China A-Stock Returns

Abstract

The theory of quadratic variation suggests that, realized volatility is an unbiased and highly efficient estimator of return volatility under suitable conditions. In this article, we compare the realized logarithmic volatilities models VAR-RV and AR-RV computed from high-frequency intra-period data with the traditional daily return evaluation models VAR-R and Daily-GARCH in China A-stock market. The result suggests that the realized volatility do a better and more efficient measure in evaluating and forecasting the volatility characteristic for China stock market.