Multi-Step-Ahead Prediction with Neural Networks
We review existing approaches in using neural networks for solving multi-stepahead prediction problems. A few experiments allow us to further explore the relationship between the ability to learn longer-range dependencies and performance in multi-step-ahead prediction. We eventually focus on characteristics of various multi-step-ahead prediction problems that encourage us to prefer one method over another.
Romuald BONÉ, Michel CRUCIANU
Time Series Prediction, Neural Networks, Multi-step-ahead Prediction, Longrange Dependencies