A time series can often be characterized using machine learning techniques, which require feature vectors as input. The quality of the feature vectors reflects the accuracy of the utilized machine learning techniques. We propose a method for combining features extracted from two popular...

This study proposes a novel loss function for neural network models that explores the topological structure of stock realized volatility (RV) data by adding Wasserstein Distance (WD). The study shows that the proposed loss statistically significantly improves the forecast accuracy of neural...