High-Frequency-Based Volatility Model with Network Structure

This paper introduces a novel multi-variate volatility model that can accommodate appropriately defined network structures based on low-frequency and high-frequency data. The model offers substantial reductions in the number of unknown parameters and computational complexity. The model formulation, along with iterative multi-step-ahead forecasting and targeting parameterization are discussed. Quasi-likelihood functions for parameter estimation are proposed and their asymptotic properties are established. A series of simulation studies are carried out to assess the performance of parameter estimation in finite samples. Furthermore, a real data analysis demonstrates that the proposed model outperforms the existing volatility models in prediction of future variances of daily return and realized measures.