Daily Semiparametric GARCH Model Estimation Using Intraday High-Frequency Data

The GARCH model is one of the most influential models for characterizing and predicting fluctuations in economic financial studies. However, traditional commonly use daily frequency data to predict return, correlation, risk indicator assets, without taking with other frequencies into account. Hence, market information may not be sufficiently applied estimation GARCH-type models. To partially solve this problem, paper introduces intraday high-frequency improve volatility function a semiparametric model. achieve objective, proxy was proposed, which includes both symmetric asymmetric cases. Under mild conditions, asymptotic normality estimators established. Furthermore, we also discuss impact different proxies on precision. Both simulation empirical results showed that could improved by introduction data.