Modelling and Predicting the Conditional Variance of Bitcoin Daily Returns: Comparsion of Markov Switching GARCH and SV Models

TL;DR

This study compares Markov Switching-GARCH and Stochastic Autoregressive Volatility models to forecast Bitcoin's daily return volatility, finding SARV models generally outperform MS-GARCH models in predictive accuracy.

Contribution

It introduces a two-stage estimation approach and demonstrates the superior performance of SARV models over MS-GARCH models for Bitcoin volatility forecasting.

Findings

SARV models outperform MS-GARCH in forecasting Bitcoin volatility

Simple GARCH models can outperform Markov-Switching GARCH in certain cases

Two-stage estimation improves model performance

Abstract

This paper introduces a unique and valuable research design aimed at analyzing Bitcoin price volatility. To achieve this, a range of models from the Markov Switching-GARCH and Stochastic Autoregressive Volatility (SARV) model classes are considered and their out-of-sample forecasting performance is thoroughly examined. The paper provides insights into the rationale behind the recommendation for a two-stage estimation approach, emphasizing the separate estimation of coefficients in the mean and variance equations. The results presented in this paper indicate that Stochastic Volatility models, particularly SARV models, outperform MS-GARCH models in forecasting Bitcoin price volatility. Moreover, the study suggests that in certain situations, persistent simple GARCH models may even outperform Markov-Switching GARCH models in predicting the variance of Bitcoin log returns. These findings offer valuable guidance for risk management experts, highlighting the potential advantages of SARV models in managing and forecasting Bitcoin price volatility.