Forecasting bitcoin volatility: Evidence from the options market

Abstract

This paper studies a large number of bitcoin (BTC) options traded on the options exchange Deribit. We use the trades to calculate implied volatility (IV) and analyze if volatility forecasts can be improved using such information. IV is less accurate than AutoRegressive–Moving‐Average or Heterogeneous Auto‐Regressive model forecasts in predicting short‐term BTC volatility (1 day ahead), but superior in predicting long‐term volatility (7, 10, 15 days ahead). Furthermore, a combination of IV and model‐based forecasts provides the highest accuracy for all forecasting horizons revealing that the BTC options market contains unique information.

Forecasting Bitcoin Futures: A Lasso-BMA Two-Step Predictor Selection for Investment and Hedging Strategies

“…Shynkevich (2020) concludes that the introduction of Bitcoin futures disrupts the predictive power of technical trading rules. Hoang and Baur (2020) calculate implied volatility from a number of Bitcoin options traded on the options exchange Deribit and show that this predictor is superior in predicting 7- to 15-day than 1-day ahead realized volatility. Lee et al (2020) discover that the Bitcoin futures basis provides some predictive power for spot price and risk premium changes.…”

Forecasting realized volatility of Bitcoin: The informative role of price duration

“…A recurring question, in equity‐related empirical literature, is whether option markets play a role in providing information for volatility forecasts (see Slim et al, 2020, and references therein). In the context of the Bitcoin's derivative market, Hoang and Baur (2020) demonstrate that implied volatility provides incremental information about future volatility beyond that contained in GARCH and HAR‐type models. More closely related to our findings, Gkillas et al (2021) report a significant incremental forecasting value of including on‐chain transaction activity, measured by the total number of unique transactions recorded on the blockchain, into the HAR model.…”

Transfer‐entropy‐based dynamic feature selection for evaluating Bitcoin price drivers

“…Moreover, these characteristics of BTC price time series, along with the lack of a known relationship between BTC price fluctuations and the movements in other financial time series, hinder a consensus on BTC price drivers. Despite the lack of consensus on BTC price drivers, forecasters have conducted parallel investigations to discover a forecasting model and features for BTC price (Chen et al, 2021; Kurbucz, 2019; Mallqui & Fernandes, 2019; Parvini et al, 2020), return (Akyildirim et al, 2020; Atsalakis et al, 2019; Balcilar et al, 2017; Catania et al, 2019; Parvini et al, 2022), and volatility (Hoang & Baur, 2020; Peng et al, 2018; Trucíos, 2019). Improving BTC forecasting models is important because more accurate estimates of the future values of BTC time series help portfolio managers to better formulate risk/hedging strategies and aid traders to design more profitable trading strategies.…”

“…, return (Akyildirim et al, 2020;Atsalakis et al, 2019;Balcilar et al, 2017;Catania et al, 2019;Parvini et al, 2022), and volatility (Hoang & Baur, 2020;Peng et al, 2018;Trucíos, 2019). Improving BTC forecasting models is important because more accurate estimates of the future values of BTC time series help portfolio managers to better formulate risk/hedging strategies and aid traders to design more profitable trading strategies.…”