FORECASTING OF BITCOIN CRYPTOCURRENCY PRICE USING THE AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) METHOD
Nella Cornelya (1*), Dian Cahyawati (2), Ning Eliyati (3), Robinson Sitepu (4), Eka Susanti (5), Novi Rustiana Dewi (6), Putra BJ Bangun (7)

Departement of Mathematics, Faculty of Mathematics and Natural Sciences,
Sriwijaya University

Abstract

Bitcoin is a type of cryptocurrency that has the largest market capitalization as a digital investment tool. Bitcoin prices often fluctuate, making it necessary to forecast in order to help investors in predicting the profit of their digital investment. One of the methods that can be used for predicting the price of Bitcoin based on the daily preices is Autoregressive Integrated Moving Average (ARIMA). The aim of this research is to create the best ARIMA model and predict the current price of Bitcoin. The data used in this research is secondary data obtained from websites supervised by the Commodity Futures Trading Supervisory Agency starting from 1 January 2023 to 30 September 2023. The research results show that the ARIMA model (0,2,1) provides a MAPE value of 3.17%, which means the ARIMA model^s forecasting ability is very good.

Keywords: ARIMA, Bitcoin, Cryptocurrency, Forecasting.

Topic: Mathematics and Its Applications