Forecasting High-Dimensional Realized Volatility Matrices Using A Factor Model

TL;DR

This paper introduces a factor model with a diagonal CAW component for high-dimensional realized volatility matrices, effectively reducing parameters while maintaining forecasting accuracy in financial applications.

Contribution

The paper proposes a novel factor model with a diagonal CAW structure that addresses the curse of dimensionality in high-dimensional covariance matrix forecasting.

Findings

Significant reduction in the number of parameters.

Model achieves comparable performance to benchmark VAR models.

Asymptotic theory established for parameter estimation.

Abstract

Modeling and forecasting covariance matrices of asset returns play a crucial role in finance. The availability of high frequency intraday data enables the modeling of the realized covariance matrix directly. However, most models in the literature suffer from the curse of dimensionality. To solve the problem, we propose a factor model with a diagonal CAW model for the factor realized covariance matrices. Asymptotic theory is derived for the estimated parameters. In an extensive empirical analysis, we find that the number of parameters can be reduced significantly. Furthermore, the proposed model maintains a comparable performance with a benchmark vector autoregressive model.