Monitoring Financial Stability Based on Prediction of Cryptocurrencies Price Using Intelligent Algorithm

Abstract

Financial stability is the problem that correlated with many aspects. In this digital era, virtual currency has emerged as one of financial assets. Fluctuation from cryptocurrencies price made an impact into financial stability indirectly. It has been proved by the market capitalization of publicly traded cryptocurrencies in 2019 reach USD 240 billion. The percentages reach 66% for the market capitalization. Three highest cryptocurrencies uphold are Bitcoin, Ethereum and XRP. This condition made these three cryptocurrencies become the important investment products. However, cryptocurrency is the product with high volatility. Because of that, this study aims to monitor financial stability from cryptocurrencies prediction using artificial intelligent algorithm. This research copes the problem by predicting bitcoin, Ethereum and XRP using three different intelligent algorithms, which are K-Nearest Neighbours (KNN), Support Vector Machine (SVM) and Long Short-Term Memory (LSTM). By this prediction had been figured out about up and down value of cryptocurrencies using Root Mean Square Error (RMSE) to evaluate stabilization of finance. Accuration system with LSTM shown that the price of cryptocurrency will fit with the data actual using LSTM, in which the accuracy around 80%. This condition meant that LSTM had been succeeded to proposed as algorithm that could fit the cryptocurrencies value. It could strengthen usability to indicate financial stability refer into cryptocurrencies price.