Analysing the distribution properties of Bitcoin returns

Abstract

This study exploits several conditional heteroskedasticity models with various supported distributions in order to find the best distribution as well as the best GARCH-type model that may be used to model volatility of Bitcoin returns. Innovatively, the study is able to establish that pre-testing the residuals of Bitcoin returns for the best distribution can help to identify the appropriate distribution when modelling with GARCH-type models regardless of the data frequency.