Dynamic covariance modeling with artificial neural networks

Wing Ki LIU, Mike K. P. SO, Man Ying Amanda CHU

Research output: Contribution to journalArticlespeer-review

Abstract

This article proposes a novel multivariate generalized autoregressive conditionally heteroscedastic (GARCH) model that incorporates the modified Cholesky decomposition for a covariance matrix in order to reduce the number of covariance parameters and increase the interpretation power of the model. The modified Cholesky decomposition for covariance matrix reduces the number of covariance parameters to p(p+1)/2, where p is the dimension of the stocks in the data, and enables us to obtain a regression equation. To account for the nonlinearity in the GARCH model, the parameters in our model are modeled using long short-term memory. The proposed model is compared with DCC model with respect to portfolio optimization and the distances between the actual covariance matrices and predicted covariance matrices. It is found that although the distances may or may not be reduced by our proposed model in different cases presented in this article, our proposed model outperforms the DCC model in terms of mean portfolio returns. Copyright © 2021 Informa UK Limited .
Original languageEnglish
Pages (from-to)15-42
JournalCommunications in Statistics: Case Studies, Data Analysis and Applications
Volume8
Issue number1
Early online date15 Sep 2021
DOIs
Publication statusPublished - 2022

Citation

Liu, W. K., So, M. K. P., & Chu, A. M. Y. (2022). Dynamic covariance modeling with artificial neural networks. Communications in Statistics: Case Studies, Data Analysis and Applications, 8(1), 15-42. doi: 10.1080/23737484.2021.1972876

Keywords

  • Conditional heteroscedasticity
  • Dynamic conditional correlation
  • GARCH
  • Long short-term memory
  • Volatility modeling

Fingerprint

Dive into the research topics of 'Dynamic covariance modeling with artificial neural networks'. Together they form a unique fingerprint.