Hierarchically Coherent Multivariate Mixture Networks

Kin G. Olivares, David Luo, Cristian Challu, Stefania La Vattiata, Max Mergenthaler +1 more
5/11/2023
stat.MLcs.LGstat.ME

Official code: Nixtla/neuralforecast (4,219 stars, Apache-2.0, PyTorch, last commit 2026-04-30).

Abstract

Large collections of time series data are often organized into hierarchies with different levels of aggregation; examples include product and geographical groupings. Probabilistic coherent forecasting is tasked to produce forecasts consistent across levels of aggregation. In this study, we propose to augment neural forecasting architectures with a coherent multivariate mixture output. We optimize the networks with a composite likelihood objective, allowing us to capture time series' relationships while maintaining high computational efficiency. Our approach demonstrates 13.2% average accuracy improvements on most datasets compared to state-of-the-art baselines. We conduct ablation studies of the framework components and provide theoretical foundations for them. To assist related work, the code is available at this https://github.com/Nixtla/neuralforecast.

View on arXivView PDF

Code Implementations(1)

Nixtla/neuralforecastOfficial100%
PyTorchApache-2.0

Scalable and user friendly neural :brain: forecasting algorithms.

4,219500Shell, PythonApr 26, 20215 months agoApache-2.0
baselinesbaselines-zoodeep-learningdeep-neural-networksdeepar+15 more

Similar Papers

$\clubsuit$ CLOVER $\clubsuit$: Probabilistic Forecasting with Coherent Learning Objective Reparameterization

85% match2023

Learning Mixture Structure on Multi-Source Time Series for Probabilistic Forecasting

83% match2023

CoRe: Coherency Regularization for Hierarchical Time Series

81% match2025

Are KANs Effective for Multivariate Time Series Forecasting?

80% match1 repo1192024

Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting

80% match2023

Hierarchical Time Series Forecasting Via Latent Mean Encoding

79% match1 repo2252025

Cite this paper

@article{olivares2023hierarchically,
  title  = {Hierarchically Coherent Multivariate Mixture Networks},
  author = {Kin G. Olivares and David Luo and Cristian Challu and Stefania La Vattiata and Max Mergenthaler and Artur Dubrawski},
  year   = {2023},
  eprint = {2305.07089},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.07089v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.07089.svg)](https://codeofpaper.com/paper/2305.07089)

Discussion