A hybrid model for intraday volatility prediction in Bitcoin markets

Prakash Raj, Koushik Bera and N. Selvaraju

The North American Journal of Economics and Finance, 2025, vol. 78, issue C

Abstract: Volatility modeling in cryptocurrencies poses unprecedented challenges due to extreme price fluctuation, 24/7 trading cycles, and decentralized and speculative environments. This article presents a novel hybrid BEMD-REGARCH model by integrating the bivariate empirical mode decomposition (BEMD) with the realized exponential generalized autoregressive conditional heteroscedasticity (REGARCH) model to estimate the volatility of cryptocurrencies. The highlights include the use of intraday hourly returns and realized variance, and the model forecasts intraday 1-hour-ahead volatility. Testing the hybrid model on various datasets ensures robustness, and the model yields superior volatility forecasting gains over the traditional REGARCH model on various performance metrics. In addition, BEMD trumps EMD by scoring lower forecasting errors than the EMD-GARCH model. In summary, applying BEMD to the REGARCH model enhances its forecasting performance.

Keywords: Decomposition algorithm; EMD; REGARCH; BEMD; Intraday; Volatility forecasting (search for similar items in EconPapers)
JEL-codes: C52 C58 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S106294082500066X
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:ecofin:v:78:y:2025:i:c:s106294082500066x

DOI: 10.1016/j.najef.2025.102426

Access Statistics for this article

The North American Journal of Economics and Finance is currently edited by Hamid Beladi

More articles in The North American Journal of Economics and Finance from Elsevier
Bibliographic data for series maintained by Catherine Liu ().