For2For: Learning to forecast from forecasts
This approach improves forecasting accuracy for time series analysis, particularly in competitive benchmarks, but is incremental as it builds on existing ensemble methods.
The paper tackles time series forecasting by using forecasts from standard methods as inputs to a machine learning model, achieving top performance on quarterly series and near-top on monthly series in the M4 competition dataset.
This paper presents a time series forecasting framework which combines standard forecasting methods and a machine learning model. The inputs to the machine learning model are not lagged values or regular time series features, but instead forecasts produced by standard methods. The machine learning model can be either a convolutional neural network model or a recurrent neural network model. The intuition behind this approach is that forecasts of a time series are themselves good features characterizing the series, especially when the modelling purpose is forecasting. It can also be viewed as a weighted ensemble method. Tested on the M4 competition dataset, this approach outperforms all submissions for quarterly series, and is more accurate than all but the winning algorithm for monthly series.