Using Machine Learning Techniques to Combine Forecasting Methods

We present here an original work that uses machine learning techniques to combine time series forecasts. In this proposal, a machine learning technique uses features of the series at hand to define adequate weights for the individual forecasting methods being combined. The combined forecasts are the weighted average of the forecasts provided by the individual methods. In order to evaluate this solution, we implemented a prototype that uses a MLP network to combine two widespread methods. The experiments performed revealed significantly accurate forecasts.