State Heterogeneity Analysis of Financial Volatility using high?frequency Financial Data

Recently, to account for low-frequency market dynamics, several volatility models, employing high-frequency financial data, have been developed. However, in markets, we often observe that processes depend on economic states, so they a state heterogeneous structure. In this paper, study dynamics based introduce novel model continuous Ito diffusion process whose intraday instantaneous evolves depending the exogenous variable, as well its integrated volatility. We call it GARCH-Ito (SG-Ito) model. suggest quasi-likelihood estimation procedure with realized proxy and establish asymptotic behaviors. Moreover, test heterogeneity, develop Wald test-type hypothesis testing procedure. The results of empirical studies existence leverage, investor attention, illiquidity, stock comovement, post-holiday effect S&P 500 index