Multi-output Ensembles for Multi-step Forecasting
This work addresses a gap in forecasting methods for multi-step tasks, though it is incremental as it builds on existing ensemble techniques.
The paper tackled the problem of applying dynamic ensembles to multi-step ahead forecasting, finding that ensembles based on arbitrating and windowing performed best according to average rank, but most approaches struggled to outperform a static ensemble with equal weights as the horizon increased.
This paper studies the application of ensembles composed of multi-output models for multi-step ahead forecasting problems. Dynamic ensembles have been commonly used for forecasting. However, these are typically designed for one-step-ahead tasks. On the other hand, the literature regarding the application of dynamic ensembles for multi-step ahead forecasting is scarce. Moreover, it is not clear how the combination rule is applied across the forecasting horizon. We carried out extensive experiments to analyze the application of dynamic ensembles for multi-step forecasting. We resorted to a case study with 3568 time series and an ensemble of 30 multi-output models. We discovered that dynamic ensembles based on arbitrating and windowing present the best performance according to average rank. Moreover, as the horizon increases, most approaches struggle to outperform a static ensemble that assigns equal weights to all models. The experiments are publicly available in a repository.