Multivariate range-based EGARCH models

• Published in: International review of financial analysis Vol. 100; p. 103983
• Main Authors: Yan, Lili, Kellard, Neil M., Lambercy, Lyudmyla
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.04.2025
• Subjects: DCC; EGARCH; EWMA; Portfolio modelling; Range-based covariance forecasting; Range-based covariance forecasting; C53; DCC; EWMA; Portfolio modelling; G11; C58; EGARCH
• ISSN: 1057-5219
• DOI: 10.1016/j.irfa.2025.103983

The dynamic conditional correlation (DCC) and co-range models are two main frameworks used to incorporate range-based univariate volatility. Using the two approaches, we construct novel multivariate range-based EGARCH (REGARCH) models: a DCC-REGARCH and co-range REGARCH (CRREGARCH) model, and a co-range CARR (CRCARR) model. We compare these models with five existing models over twelve forecast horizons, ranging from one to twelve weeks, covering currencies and ETFs. Among the eight models, the DCC-REGARCH and CRREGARCH models show the best performance in out-of-sample forecasting of the variance-covariance matrix across a range of market conditions and forecast horizons. These models also generate the lowest variance and turnover for global minimum-variance (GMV) portfolios in the majority of cases. •Two multivariate REGARCH models are built based on the DCC and corange frameworks.•Two data sets, currencies and ETFs are employed to assess the developed models.•The two models are evaluated over twelve forecasting horizons.•The developed models outperform the competing range-based models.