A Trainable Reconciliation Method for Hierarchical Time-Series
This method offers improved accuracy for businesses and organizations that rely on hierarchical demand forecasting, such as in supply chain management, by ensuring consistency across different aggregation levels.
This paper addresses the problem of reconciling hierarchical time-series forecasts, where independent forecasts at different levels do not sum consistently. The authors propose a new encoder-decoder neural network-based reconciliation strategy that consistently matches or exceeds the performance of existing methods across four real-world datasets.
In numerous applications, it is required to produce forecasts for multiple time-series at different hierarchy levels. An obvious example is given by the supply chain in which demand forecasting may be needed at a store, city, or country level. The independent forecasts typically do not add up properly because of the hierarchical constraints, so a reconciliation step is needed. In this paper, we propose a new general, flexible, and easy-to-implement reconciliation strategy based on an encoder-decoder neural network. By testing our method on four real-world datasets, we show that it can consistently reach or surpass the performance of existing methods in the reconciliation setting.