End-to-End Learning of Coherent Probabilistic Forecasts for Hierarchical Time Series
Abstract
This paper presents a novel approach for hierarchical time series forecasting that produces coherent, probabilistic forecasts without requiring any explicit post-processing reconciliation. Unlike the state-of-the-art, the proposed method simultaneously learns from all time series in the hierarchy and incorporates the reconciliation step into a single trainable model. This is achieved by applying the reparameterization trick and casting reconciliation as an optimization problem with a closed-form solution. These model features make end-to-end learning of hierarchical forecasts possible, while accomplishing the challenging task of generating forecasts that are both probabilistic and coherent. Importantly, our approach also accommodates general aggregation constraints including grouped and temporal hierarchies. An extensive empirical evaluation on real-world hierarchical datasets demonstrates the advantages of the proposed approach over the state-of-the-art.
Additional Information
© 2021 by the author(s). The authors thank Souhaib Ben Taieb for insightful discussions on this topic and his timely help in reproducing results with PERMBU-MINT. The authors are also grateful for being able to build on the work of Valentin Flunkert and David Salinas in both concepts and code.Attached Files
Published - rangapuram21a.pdf
Supplemental Material - rangapuram21a-supp.pdf
Files
Name | Size | Download all |
---|---|---|
md5:3e6dc98ca7fae4b75712bf38c7cc0ab2
|
179.7 kB | Preview Download |
md5:7287642cea22486312ccc3212207a34e
|
1.5 MB | Preview Download |
Additional details
- Eprint ID
- 115236
- Resolver ID
- CaltechAUTHORS:20220622-211540696
- Created
-
2022-06-28Created from EPrint's datestamp field
- Updated
-
2022-06-28Created from EPrint's last_modified field