Forecasting Digital Asset Return: An Application of Machine Learning Model DOI Creative Commons
Vito Ciciretti, Alberto Pallotta, Suman Lodh

et al.

International Journal of Finance & Economics, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

ABSTRACT In this study, we aim to identify the machine learning model that can overcome limitations of traditional statistical modelling techniques in forecasting Bitcoin prices. Also, outline necessary conditions make suitable. We draw on a multivariate large data set prices and its market microstructure variables apply three models, namely double deep Q‐learning, XGBoost ARFIMA‐GARCH. The findings show Q‐learning outperforms others terms returns Sortino ratio is capable one‐step‐ahead sign forecast even synthetic data. These critical insights literature will support practitioners regulators an economically viable cryptocurrency return model.

Language: Английский

Exploiting synthetic data generation to enhance pollution prediction DOI
Juan Morales-García, Emilio Ramos-Sorroche, Sara Balderas-Díaz

et al.

Applied Soft Computing, Journal Year: 2025, Volume and Issue: unknown, P. 113076 - 113076

Published: April 1, 2025

Language: Английский

Citations

0
Forecasting Digital Asset Return: An Application of Machine Learning Model DOI Creative Commons
Vito Ciciretti, Alberto Pallotta, Suman Lodh

et al.

International Journal of Finance & Economics, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

ABSTRACT In this study, we aim to identify the machine learning model that can overcome limitations of traditional statistical modelling techniques in forecasting Bitcoin prices. Also, outline necessary conditions make suitable. We draw on a multivariate large data set prices and its market microstructure variables apply three models, namely double deep Q‐learning, XGBoost ARFIMA‐GARCH. The findings show Q‐learning outperforms others terms returns Sortino ratio is capable one‐step‐ahead sign forecast even synthetic data. These critical insights literature will support practitioners regulators an economically viable cryptocurrency return model.

Language: Английский

Citations

0