Predicting the Price of Bitcoin Using Hybrid ARIMA and Machine Learning

摘要

Bitcoin is one of the most popular cryptocurrencies in the world, has attracted broad interests from researchers in recent years. In this work, Autoregressive Integrate Moving Average (ARIMA) model and machine learning algorithms will be implemented to predict the closing price of Bitcoin the next day. After that, we present hybrid methods between ARIMA and machine learning to improve prediction of Bitcoin price. Experiment results showed that hybrid methods have improved accuracy of predicting through RMSE and MAPE.