Time Series Prediction

We have the world’s first prediction for multivariate time series prediction based on deep convolutional neural networks (CNN) and recurrent neural networks (LSTM). We’ll apply this technology to different domains and markets; the European Energy Market (EPEX), the S&P500 Index, Forex, and trading of commodities.


The network use an architecture that allow them to learn useful features from the time series history as well as from other data sources like for example market data and indices, expert predictions, and sentiment data.