Fractional differentiation and its use in machine learning

Abstract This article covers the implementation of fractional (non-integer order) differentiation on real data four datasets based stock prices main international indexes: WIG 20, S&P 500, DAX and Nikkei 225. concept has been proposed by Lopez de Prado [5] to find most appropriate balance between zero fully differentiated time series. The aim is making series stationary while keeping its memory predictive power. In addition, this paper compares classical in terms effectiveness artificial neural networks. Root mean square error (RMSE) absolute (MAE) are employed comparison. Our investigations have determined conclusion that plays an important role leads more accurate predictions case ANN.