Predicting the Trends of Price for Ethereum Using Deep Learning Techniques

Abstract

This study intends to predict the trends of price for a cryptocurrency, i.e. Ethereum based on Deep Learning techniques considering its trends on time series particularly. This study analyses as how Deep Learning techniques such as Multi-layer perceptron (MLP) and Long Short-Term Memory (LSTM) help in predicting the price trends of Ethereum. These techniques have been applied based on historical data that were computed per day, hour, and minute wise. The data set is sourced from the CoinDesk repository. The performance of the obtained models are critically assessed using statistical indicators like Mean Absolute Error (MAE), Mean Squared Error (MSE) and Root Mean Squared Error (RMSE).