Journal article 337 views 204 downloads
Prediction and decoding of metaverse coin dynamics: a granular quest using MODWT-Facebook’s prophet-TBATS and XAI methodology
,
Kazim Topuz
Annals of Operations Research, Volume: 346, Issue: 3, Pages: 2423 - 2459
Swansea University Author:
Mohammad Abedin
-
PDF | Version of Record
Open Access. This article is licensed under a Creative Commons Attribution 4.0 International License (CC-BY 4.0).
Download (4.4MB)
DOI (Published version): 10.1007/s10479-025-06491-1
Abstract
The growing media buzz and industry focus on the emergence and rapid development of Metaverse technology have paved the way for the escalation of multifaceted research. Specific Metaverse coins have come into existence, but they have barely seen any traction among practitioners despite their tremend...
| Published in: | Annals of Operations Research |
|---|---|
| ISSN: | 0254-5330 1572-9338 |
| Published: |
Springer Nature
2025
|
| Online Access: |
Check full text
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa68745 |
| Abstract: |
The growing media buzz and industry focus on the emergence and rapid development of Metaverse technology have paved the way for the escalation of multifaceted research. Specific Metaverse coins have come into existence, but they have barely seen any traction among practitioners despite their tremendous potential. The current work endeavors to deeply analyze the temporal characteristics of 6 Metaverse coins through the lens of predictive analytics and explain the forecasting process. The dearth of research imposes serious challenges in building the forecasting model. We resort to a granular prediction setup incorporating the Maximal Overlap Discrete Wavelet Transformation (MODWT) technique to disentangle the original series into subseries. Facebook's Prophet and TBATS algorithms are utilized to individually draw predictions on granular components. Aggregating components-wise forecasted figures achieve the final forecast. Facebook's Prophet is deployed in a multivariate setting, applying a set of explanatory features covering macroeconomic, technical, and social media indicators. Rigorous performance checks justify the efficiency of the integrated forecasting framework. Additionally, to interpret the black box typed prediction framework, two explainable artificial intelligence (XAI) frameworks, SHAP and LIME, are used to gauge the nature of the influence of the predictor variables, which serve several practical insights. |
|---|---|
| Keywords: |
Metaverse coin; Maximal overlap discrete wavelet transformation (MODWT); Facebook’s prophet; TBATS; Explainable artificial intelligence (XAI) |
| College: |
Faculty of Humanities and Social Sciences |
| Funders: |
Swansea University |
| Issue: |
3 |
| Start Page: |
2423 |
| End Page: |
2459 |