Corresponding Author ORCID Identifer
Sew Lai Ng https://orcid.org/0000-0002-8610-581X
Document Type
Research
Abstract
This study examines whether realised range volatility improves multivariate volatility forecasting performance in an emerging equity market setting. A comparison between vector heterogeneous autoregressive models constructed with realised volatility, realised range volatility, and conventional return-based volatility measures is conducted using sectoral data from the Bursa Malaysia. High-frequency data from September 2018 to December 2024 were collected from the financial data provider Bloomberg. The analysis evaluates out-of-sample forecasting accuracy using a set of 5 loss functions, including symmetric (MSE and MAE) and asymmetric (QLIKE) criteria. The outcomes provide compelling and strong evidence that realised range volatility measures deliver superior forecasting performance across sectoral pairs. Realised range volatility models consistently dominate alternatives relying on realised volatility and daily returns, underscoring the significance of efficient and effective volatility measurement in multivariate frameworks. These findings have direct implications for portfolio risk management, regulatory capital assessment, and volatility modelling in emerging markets featured by microstructure frictions and sectoral interdependence.
Keywords
VHAR, Realised-Range Volatility, Multivariate Volatility Forecasting, High Frequency Data, Sectoral Markets, Bursa Malaysia
Cite This Article
Fadhlur, R. A., Nursyahwani, M. S., Maciej, W., Haslinda, M., (2026). Pricing under pressure: Cost volatility, customer affordability and value perception in micro and small food businesses. International Journal of Management, Finance and Accounting, 7(2), 387–425. https://doi.org/10.33093/ijomfa.2026.7.2.12
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
