Hierarchical Time Series Forecasting Via Latent Mean Encoding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Salatiello, Alessandro, Birr, Stefan, Kunz, Manuel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913910333898752
author Salatiello, Alessandro
Birr, Stefan
Kunz, Manuel
author_facet Salatiello, Alessandro
Birr, Stefan
Kunz, Manuel
contents Coherently forecasting the behaviour of a target variable across both coarse and fine temporal scales is crucial for profit-optimized decision-making in several business applications, and remains an open research problem in temporal hierarchical forecasting. Here, we propose a new hierarchical architecture that tackles this problem by leveraging modules that specialize in forecasting the different temporal aggregation levels of interest. The architecture, which learns to encode the average behaviour of the target variable within its hidden layers, makes accurate and coherent forecasts across the target temporal hierarchies. We validate our architecture on the challenging, real-world M5 dataset and show that it outperforms established methods, such as the TSMixer model.
format Preprint
id arxiv_https___arxiv_org_abs_2506_19633
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hierarchical Time Series Forecasting Via Latent Mean Encoding
Salatiello, Alessandro
Birr, Stefan
Kunz, Manuel
Machine Learning
Artificial Intelligence
Coherently forecasting the behaviour of a target variable across both coarse and fine temporal scales is crucial for profit-optimized decision-making in several business applications, and remains an open research problem in temporal hierarchical forecasting. Here, we propose a new hierarchical architecture that tackles this problem by leveraging modules that specialize in forecasting the different temporal aggregation levels of interest. The architecture, which learns to encode the average behaviour of the target variable within its hidden layers, makes accurate and coherent forecasts across the target temporal hierarchies. We validate our architecture on the challenging, real-world M5 dataset and show that it outperforms established methods, such as the TSMixer model.
title Hierarchical Time Series Forecasting Via Latent Mean Encoding
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.19633