Using Advanced Machine Learning Techniques to Predict the Sales Volume of Non-Fungible Tokens

Ozge Camalan (), Sahika Gokmen () and Sibel Atan ()
Additional contact information
Ozge Camalan: Atılım University Department of Economics
Sahika Gokmen: Hacı Bayram Veli University, Department of Econometrics and Uppsala University, Department of Statistics
Sibel Atan: Hacı Bayram Veli University, Department of Econometrics

World Journal of Applied Economics, 2024, vol. 10, issue 1, 17-27

Abstract: Non-fungible tokens (NFTs) are a type of digital asset based on blockchain that contain unique codes verifying the authenticity and ownership of different assets such as art pieces, music, gaming items, collections, and so on. This phenomenon and its markets have grown significantly since the beginning of 2021. This study, using daily data between November 2017 and November 2022, predicts the volume of NFT sales by utilising Random Forest (RF), GBM, XGBoost, and LightGBM methods from the community machine learning methods. In the predictions, several financial variables, including Gold, Bitcoin/USD, Ethereum/USD, S&P 500 index, Nasdaq 100, Oil/USD, Euro/USD, and CDS data, are treated as independent variables. According to the results, XGBoost is found to be the best prediction method for NFT market volume estimation concerning several statistical criteria, e.g., MAE, MAPE, and RMSE, and the most significant influential feature in determining prices is the Ethereum/USD exchange rate.

Keywords: Financial assets; Non-fungible tokens; Machine learning (search for similar items in EconPapers)
JEL-codes: C53 G12 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journal.econworld.org/index.php/econworld/article/view/242/90 (application/pdf)

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:ana:journl:v:10:y:2024:i:1:p:17-27

DOI: 10.22440/wjae.10.1.2

Access Statistics for this article

World Journal of Applied Economics is currently edited by Unal Tongur

More articles in World Journal of Applied Economics from WERI-World Economic Research Institute Contact information at EDIRC.
Bibliographic data for series maintained by Unal Tongur ().