Forecasting method for Ethereum cryptocurrency with autoregressive integrated moving average (ARIMA) model using high frequency data

Ethereum is a type of cryptocurrency that exhibits an open-source, public, blockchain-based distributed computing platform and operating system featuring smart contract functionality. Ethereum is a cryptocurrency whose blockchain is generated by the Ethereum platform. Ethereum can be transferred between accounts and used to compensate participant mining nodes for computations performed. Ethereum attracts the attention of investors when the price of ether is increased sharply in the year of 2017. Therefore, the objective of this study is to forecast the value of the exchange rate of Ethereum using high-frequency data. The findings of this study will help investors to evaluate the appropriateness of Ethereum to be selected as one profile in the investment portfolio. This study developed the forecasting method using the autoregressive integrated moving average (ARIMA). The forecasting model investigated in this study is ARIMA (1,1,1) and ARIMA(1,1,2). ARIMA (1,1,2) shows a higher value of R-squared and lower the value of Akaike info criterion. Therefore, ARIMA(1,1,2) is a better model to estimate the likelihood of a model to estimate the future values of the Ethereum exchange rate. The ex-ante forecasting calculation indicates the mean absolute percentage error is 0.54 percentages. Therefore, the ARIMA (1,1,2) is a reliable model for forecasting. In conclusion, the Ethereum exchange rate can be forecast using high-frequency data with the implementation of reliable ARIMA model namely ARIMA (1,1,2).