Coherent Probabilistic Forecasting of Temporal Hierarchies

Syama Sundar Rangapuram, Shubham Kapoor, Rajbir Singh Nirwan, Pedro Mercado, Tim Januschowski, Yuyang Wang, Michael Bohlke-Schneider
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:9362-9376, 2023.

Abstract

Forecasts at different time granularities are required in practice for addressing various business problems starting from short-term operational to medium-term tactical and to long-term strategic planning. These forecasting problems are usually treated independently by learning different ML models which results in forecasts that are not consistent with the temporal aggregation structure, leading to inefficient decision making. Some of the recent work addressed this problem, however, it uses a post-hoc reconciliation strategy, which results in sub-optimal results and cannot produce probabilistic forecasts. In this paper, we present a global model that produces coherent, probabilistic forecasts for different time granularities by learning joint embeddings for the different aggregation levels with graph neural networks and temporal reconciliation. Temporal reconciliation not only enables consistent decisions for business problems across different planning horizons but also improves the quality of forecasts at finer time granularities. A thorough empirical evaluation illustrates the benefits of the proposed method.

Cite this Paper


BibTeX
@InProceedings{pmlr-v206-rangapuram23a, title = {Coherent Probabilistic Forecasting of Temporal Hierarchies}, author = {Rangapuram, Syama Sundar and Kapoor, Shubham and Nirwan, Rajbir Singh and Mercado, Pedro and Januschowski, Tim and Wang, Yuyang and Bohlke-Schneider, Michael}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {9362--9376}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/rangapuram23a/rangapuram23a.pdf}, url = {https://proceedings.mlr.press/v206/rangapuram23a.html}, abstract = {Forecasts at different time granularities are required in practice for addressing various business problems starting from short-term operational to medium-term tactical and to long-term strategic planning. These forecasting problems are usually treated independently by learning different ML models which results in forecasts that are not consistent with the temporal aggregation structure, leading to inefficient decision making. Some of the recent work addressed this problem, however, it uses a post-hoc reconciliation strategy, which results in sub-optimal results and cannot produce probabilistic forecasts. In this paper, we present a global model that produces coherent, probabilistic forecasts for different time granularities by learning joint embeddings for the different aggregation levels with graph neural networks and temporal reconciliation. Temporal reconciliation not only enables consistent decisions for business problems across different planning horizons but also improves the quality of forecasts at finer time granularities. A thorough empirical evaluation illustrates the benefits of the proposed method.} }
Endnote
%0 Conference Paper %T Coherent Probabilistic Forecasting of Temporal Hierarchies %A Syama Sundar Rangapuram %A Shubham Kapoor %A Rajbir Singh Nirwan %A Pedro Mercado %A Tim Januschowski %A Yuyang Wang %A Michael Bohlke-Schneider %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-rangapuram23a %I PMLR %P 9362--9376 %U https://proceedings.mlr.press/v206/rangapuram23a.html %V 206 %X Forecasts at different time granularities are required in practice for addressing various business problems starting from short-term operational to medium-term tactical and to long-term strategic planning. These forecasting problems are usually treated independently by learning different ML models which results in forecasts that are not consistent with the temporal aggregation structure, leading to inefficient decision making. Some of the recent work addressed this problem, however, it uses a post-hoc reconciliation strategy, which results in sub-optimal results and cannot produce probabilistic forecasts. In this paper, we present a global model that produces coherent, probabilistic forecasts for different time granularities by learning joint embeddings for the different aggregation levels with graph neural networks and temporal reconciliation. Temporal reconciliation not only enables consistent decisions for business problems across different planning horizons but also improves the quality of forecasts at finer time granularities. A thorough empirical evaluation illustrates the benefits of the proposed method.
APA
Rangapuram, S.S., Kapoor, S., Nirwan, R.S., Mercado, P., Januschowski, T., Wang, Y. & Bohlke-Schneider, M.. (2023). Coherent Probabilistic Forecasting of Temporal Hierarchies. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:9362-9376 Available from https://proceedings.mlr.press/v206/rangapuram23a.html.

Related Material