ETH analysis and predictions utilizing deep learning

This paper attempts to provide a data analysis of cryptocurrency markets. Such markets have been developed rapidly and their volatility poses significant research challenges and justifies intensive behavior analysis. For this, we develop statistical and machine learning techniques and apply them to analyze their price variations and to generate inferences. In particular, we utilize deep learning algorithms to predict the closing price of the Ethereum cryptocurrency in a short period. The price data is accumulated from Poloniex exchange and analyzed through a Convolutional Neural Network and four types of Recurrent Neural Network including the Long Short Term Memory network, the Stacked Long Short Term Memory network, the Bidirectional Long Short Term Memory network, and the Gated Recurrent Unit network. These deep learning models are benchmarked and compared under various metrics. Our experimental data suggest that certain of the above models can be utilized to predict the Ethereum closing price in real time with promising accuracy and experimentally proven profitability.