Retrected

Abstract

There is a significant association between Bitcoin and traditional financial assets such as crude oil, gold, and the U.S. dollar. This study proposes a Dynamic Conditional Correlation GARCH model (DCC-GARCH) with artificial neutral network (ANN) approach to make a better prediction for bitcoin trading. The principle of this model lies in the fact that the DCC-GARCH model can provide some important input information, such as correlation and volatility, to the artificial neural network. This study divides the data into two periods: before the COVID-19 and during the COVID-19, and each period can separated into training set and prediction set, respectively. The training set is used to find the best ANN-DCC-GARCH model in terms of prediction error, while the prediction set check the performance of Bitcoin investment decisions.