Forecasting high-dimensional realized volatility matrices using a factor model

Keren SHEN, Jianfeng YAO, Wai Keung LI

Research output: Contribution to journalArticles

5 Citations (Scopus)

Abstract

Modelling and forecasting covariance matrices of asset returns play a crucial role in many financial fields, such as portfolio allocation and asset pricing. The availability of high-frequency intraday data enables the modelling of the realized covariance matrix directly. However, most models in the literature suffer from the curse of dimensionality, i.e. the number of parameters needed increases at the rate of the square of the number of assets. To solve the problem, we propose a factor model with a diagonal Conditional Autoregressive Wishart model for the factor realized covariance matrices. Consequently, the positive definiteness of the estimated covariance matrix is ensured with the proposed model. Asymptotic theory is derived for the estimated parameters. In the extensive empirical analysis, we find that the number of parameters can be reduced significantly; to only about one-tenth of the benchmark model. Furthermore, the proposed model maintains a comparable performance with a benchmark vector autoregressive model for different forecast horizons. Copyright © 2018 Informa UK Limited, trading as Taylor & Francis Group.
Original languageEnglish
JournalQuantitative Finance
Early online dateJun 2018
DOIs
Publication statusE-pub ahead of print - Jun 2018

Citation

Shen, K., Yao, J., & Li, W. K. (2018). Forecasting high-dimensional realized volatility matrices using a factor model. Quantitative Finance. Advance online publication. doi: 10.1080/14697688.2018.1473632

Keywords

  • High-dimension
  • High-frequency
  • Realized covariance matrices
  • Factor model
  • Wishart distribution

Fingerprint Dive into the research topics of 'Forecasting high-dimensional realized volatility matrices using a factor model'. Together they form a unique fingerprint.