Anticipating Cryptocurrency Prices Using Machine Learning

Figure 4

Schematic description of Method 2. The training set is composed of features and target (T) pairs, where features are various characteristics of all currencies, computed across the days preceding time and the target is the price of at . The features-target pairs include a single currency , for all values of included between and . The test set contains a single features-target pair: the characteristics of all currencies, computed across the days preceding time and the price of at .