Bitcoin Return Dynamics Volatility and Time Series Forecasting

Bitcoin and other cryptocurrency returns show higher volatility than equity, bond, and other asset classes. Increasingly, researchers rely on machine learning techniques to forecast returns, where different machine learning algorithms reduce the forecasting errors in a high-volatility regime. We sho...

Full description

Saved in:
Bibliographic Details
Main Authors: Punit Anand, Anand Mohan Sharan
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:International Journal of Financial Studies
Subjects:
Online Access:https://www.mdpi.com/2227-7072/13/2/108
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849433125065588736
author Punit Anand
Anand Mohan Sharan
author_facet Punit Anand
Anand Mohan Sharan
author_sort Punit Anand
collection DOAJ
description Bitcoin and other cryptocurrency returns show higher volatility than equity, bond, and other asset classes. Increasingly, researchers rely on machine learning techniques to forecast returns, where different machine learning algorithms reduce the forecasting errors in a high-volatility regime. We show that conventional time series modeling using ARMA and ARMA GARCH run on a rolling basis produces better or comparable forecasting errors than those that machine learning techniques produce. The key to achieving a good forecast is to fit the correct AR and MA orders for each window. When we optimize the correct AR and MA orders for each window using ARMA, we achieve an MAE of 0.024 and an RMSE of 0.037. The RMSE is approximately 11.27% better, and the MAE is 10.7% better compared to those in the literature and is similar to or better than those of the machine learning techniques. The ARMA-GARCH model also has an MAE and an RMSE which are similar to those of ARMA.
format Article
id doaj-art-e4a04d8bc4724e0dbab686996d3db64c
institution Kabale University
issn 2227-7072
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series International Journal of Financial Studies
spelling doaj-art-e4a04d8bc4724e0dbab686996d3db64c2025-08-20T03:27:10ZengMDPI AGInternational Journal of Financial Studies2227-70722025-06-0113210810.3390/ijfs13020108Bitcoin Return Dynamics Volatility and Time Series ForecastingPunit Anand0Anand Mohan Sharan1Department of Finance and Real Estate, School of Business, Southern Connecticut State University, New Haven, CT 06515, USAFaculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL A1B3X5, CanadaBitcoin and other cryptocurrency returns show higher volatility than equity, bond, and other asset classes. Increasingly, researchers rely on machine learning techniques to forecast returns, where different machine learning algorithms reduce the forecasting errors in a high-volatility regime. We show that conventional time series modeling using ARMA and ARMA GARCH run on a rolling basis produces better or comparable forecasting errors than those that machine learning techniques produce. The key to achieving a good forecast is to fit the correct AR and MA orders for each window. When we optimize the correct AR and MA orders for each window using ARMA, we achieve an MAE of 0.024 and an RMSE of 0.037. The RMSE is approximately 11.27% better, and the MAE is 10.7% better compared to those in the literature and is similar to or better than those of the machine learning techniques. The ARMA-GARCH model also has an MAE and an RMSE which are similar to those of ARMA.https://www.mdpi.com/2227-7072/13/2/108bitcointime seriesARMAARMA-GARCHMAERMSE
spellingShingle Punit Anand
Anand Mohan Sharan
Bitcoin Return Dynamics Volatility and Time Series Forecasting
International Journal of Financial Studies
bitcoin
time series
ARMA
ARMA-GARCH
MAE
RMSE
title Bitcoin Return Dynamics Volatility and Time Series Forecasting
title_full Bitcoin Return Dynamics Volatility and Time Series Forecasting
title_fullStr Bitcoin Return Dynamics Volatility and Time Series Forecasting
title_full_unstemmed Bitcoin Return Dynamics Volatility and Time Series Forecasting
title_short Bitcoin Return Dynamics Volatility and Time Series Forecasting
title_sort bitcoin return dynamics volatility and time series forecasting
topic bitcoin
time series
ARMA
ARMA-GARCH
MAE
RMSE
url https://www.mdpi.com/2227-7072/13/2/108
work_keys_str_mv AT punitanand bitcoinreturndynamicsvolatilityandtimeseriesforecasting
AT anandmohansharan bitcoinreturndynamicsvolatilityandtimeseriesforecasting