Since Bitcoin price is highly volatile, forecasting its volatility is crucial for many applications, such as risk management or hedging. We study which model is the most suitable for forecasting Bitcoin volatility. We consider several GARCH and two heterogeneous autoregressive (HAR) models and compare them. Since we utilize realized variance estimated from high frequency data as a proxy for true volatility, we can draw sharper conclusions than studies which use only daily data. We find that EGARCH and APARCH perform best among the GARCH models. HAR models based on realized variance perform better than GARCH models based on daily data. Superiority of HAR models over GARCH models is strongest for short-term volatility forecasts.
%0 Journal Article
%1 BERGSLI2022101540
%A Bergsli, Lykke Øverland
%A Lind, Andrea Falk
%A Molnár, Peter
%A Polasik, Michał
%D 2022
%J Research in International Business and Finance
%K ba bitcoin forecast volatility
%P 101540
%R https://doi.org/10.1016/j.ribaf.2021.101540
%T Forecasting volatility of Bitcoin
%U https://www.sciencedirect.com/science/article/pii/S0275531921001616
%V 59
%X Since Bitcoin price is highly volatile, forecasting its volatility is crucial for many applications, such as risk management or hedging. We study which model is the most suitable for forecasting Bitcoin volatility. We consider several GARCH and two heterogeneous autoregressive (HAR) models and compare them. Since we utilize realized variance estimated from high frequency data as a proxy for true volatility, we can draw sharper conclusions than studies which use only daily data. We find that EGARCH and APARCH perform best among the GARCH models. HAR models based on realized variance perform better than GARCH models based on daily data. Superiority of HAR models over GARCH models is strongest for short-term volatility forecasts.
@article{BERGSLI2022101540,
abstract = {Since Bitcoin price is highly volatile, forecasting its volatility is crucial for many applications, such as risk management or hedging. We study which model is the most suitable for forecasting Bitcoin volatility. We consider several GARCH and two heterogeneous autoregressive (HAR) models and compare them. Since we utilize realized variance estimated from high frequency data as a proxy for true volatility, we can draw sharper conclusions than studies which use only daily data. We find that EGARCH and APARCH perform best among the GARCH models. HAR models based on realized variance perform better than GARCH models based on daily data. Superiority of HAR models over GARCH models is strongest for short-term volatility forecasts.},
added-at = {2021-12-20T16:59:34.000+0100},
author = {Bergsli, Lykke Øverland and Lind, Andrea Falk and Molnár, Peter and Polasik, Michał},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2fa1cb2c67efd980aa4d32710c2c396c9/georglender},
description = {Forecasting volatility of Bitcoin - ScienceDirect},
doi = {https://doi.org/10.1016/j.ribaf.2021.101540},
interhash = {11aab6ac646eedc5b6f5f3845376f11b},
intrahash = {fa1cb2c67efd980aa4d32710c2c396c9},
issn = {0275-5319},
journal = {Research in International Business and Finance},
keywords = {ba bitcoin forecast volatility},
pages = 101540,
timestamp = {2021-12-20T15:59:34.000+0100},
title = {Forecasting volatility of Bitcoin},
url = {https://www.sciencedirect.com/science/article/pii/S0275531921001616},
volume = 59,
year = 2022
}
