Prediction accuracy improvement for Bitcoin market prices based on symmetric volatility information using artificial neural network approach

Anwar Hasan Abdullah Othman (), Salina Kassim (), Romzie Bin Rosman () and Nur Harena Binti Redzuan ()
Additional contact information
Anwar Hasan Abdullah Othman: IIUM Institute of Islamic Banking and Finance (IIiBF), International Islamic University Malaysia
Salina Kassim: IIUM Institute of Islamic Banking and Finance (IIiBF), International Islamic University Malaysia
Romzie Bin Rosman: IIUM Institute of Islamic Banking and Finance (IIiBF), International Islamic University Malaysia
Nur Harena Binti Redzuan: IIUM Institute of Islamic Banking and Finance (IIiBF), International Islamic University Malaysia

Journal of Revenue and Pricing Management, 2020, vol. 19, issue 5, No 4, 314-330

Abstract: Abstract Generally, information is the fundamental driver of assets pricing volatility in the financial market. This information can enter into the market either symmetrically or asymmetrically. The financial literature shows that Bitcoin market volatility is symmetrically informative and has a long memory to persist in the future. Additionally, the symmetricity of volatility has been revealed to be of greater sensitivity to its past values compared to the new shock of the market values. This study therefore applied the symmetric volatility structure of Bitcoin currency which can be measured through four input attributes such as open price (OP), high price (HP), low price (LP), and close price (CP) for predicting its price future trend. The study uses Rapid-Miner programme based on artificial neural network (ANN) algorithm. The optimal model employs a multilayer neural network (NN) along with an “optimised operator” with the ability to locate the optimal factor loading of the applied algorithm. The findings indicate that ANN is an effective and adequate model for correctly predicting Bitcoin market prices using symmetric volatility attributes with accuracy level of 92.15% against the actual price, whereas the low price attribute is found to be the major promoter for Bitcoin price trend with percentage of 63%. This is followed by close price, high price, and open price with percentages of 49%, 46%, and 37%, respectively. The findings of the study therefore would be a valuable and significant input for commercial purposes among the cryptocurrency market players. In other worlds, based on these outcomes investors will proactively predicate the Bitcoin price trend and make the right investment decision either to buy, hold, or sale to gain up normal market return. This is considered a pioneering study that predicates the Bitcoin price trend based on its symmetric volatility structure. As these replication findings demonstrate, the proposed model is highly promising and applicable in a real-time trading system for predicting Bitcoin price future trend and maximising investment profits in Cryptocurrency markets.

Keywords: Price predication; Bitcoin currency; Artificial neural network; Symmetric volatility information (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://link.springer.com/10.1057/s41272-020-00229-3 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

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:pal:jorapm:v:19:y:2020:i:5:d:10.1057_s41272-020-00229-3

Ordering information: This journal article can be ordered from
https://www.palgrave.com/gp/journal/41272

DOI: 10.1057/s41272-020-00229-3

Access Statistics for this article

Journal of Revenue and Pricing Management is currently edited by Ian Yeoman

More articles in Journal of Revenue and Pricing Management from Palgrave Macmillan
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().