Automated trading system using machine learning

In this study, in order to predict the next minute s closing price of Ethereum, we use six technical Indicators and the close price of BTC as inputs of several machine learning models. After that, we design an automated trading system to take short or long positions. Finally, the models evaluate in terms of performance. Results show that the performance of models can beat the buy and hold model. The random forest model has the best performance among all models with 90% accuracy. After the random forest model, the XGBoost model, decision tree, and support vector machine had the best to the weakest performance, respectively