A0932
Title: Quantile forecasts with stochastic volatility models using realized quantile measures
Authors: Yuta Yamauchi - Nagoya University (Japan) [presenting]
Genya Kobayashi - Meiji University (Japan)
Yasuhiro Omori - University of Tokyo (Japan)
Abstract: The aim is to improve the forecasting accuracy of value-at-risk (VaR) by using information on quantiles derived from high-frequency data, based on stochastic volatility models. By incorporating realized volatility, the model enhances volatility forecasting while simultaneously utilizing auxiliary information from high-frequency data related to the quantiles. A dynamic model is constructed for both the asset return quantiles and volatility as latent processes, enabling high-frequency information to be effectively incorporated through the observation equations. An empirical analysis of asset return data is conducted to assess the predictive performance of the proposed model relative to existing realized stochastic volatility models.
