| Summary: | This study provides an estimation of Bitcoin's volatility using a variation of GARCH (volatility) models. The Box-Jenkins Procedure is used throughout the analysis. The volatility clustering effect is found in Bitcoin, which suggests that GARCH models are applicable in its return series. In the empirical analysis, the standard errors of cryptocurrency returns are assumed to follow a Student-t distribution for the best fitting model. The Glosten, Jagannathan, and Runkle (GJR)- GARCH(1,1) model shows that Bitcoin's log return series exhibits an inverted leverage effect, where the volatility of Bitcoin's return tends to increase when good news happens. In financial applications, the accuracy of volatility estimation and forecasting is crucial in providing a reliable tool for risk management, option trading, asset pricing, among others. The value-at-risk measurement transforms the estimated GARCH volatility into the maximum potential loss at a certain level of confidence (95% or 99%). By including the COVID-19 period in our empirical study, we found that the pandemic has a positive impact on cryptocurrency markets. This finding provides useful information to investors in their risk management and portfolio analysis.
|
|---|