---
title: ML for hierarchical time series forecasting
url: https://www.ml-quant.com/papers/repec/eee-intfor-v-40-y-2024-i-2-p-597-615/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: RePEc:eee:intfor:v:40:y:2024:i:2:p:597-615
source_url: https://econpapers.repec.org/scripts/redir.pf?u=http%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fpii%2FS0169207022001029%3Bh%3Drepec%3Aeee%3Aintfor%3Av%3A40%3Ay%3A2024%3Ai%3A2%3Ap%3A597-615
featured: 2024-04-17
citations: unknown
topic: Econometrics & Forecasting
---


# ML for hierarchical time series forecasting

A multi-output regression model is proposed for better supply chain forecasting, using variables from different hierarchical levels to generate reliable predictions.

- Source: https://econpapers.repec.org/scripts/redir.pf?u=http%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fpii%2FS0169207022001029%3Bh%3Drepec%3Aeee%3Aintfor%3Av%3A40%3Ay%3A2024%3Ai%3A2%3Ap%3A597-615
- Identifier: RePEc:eee:intfor:v:40:y:2024:i:2:p:597-615
- Released: 2024-04-17
- First featured: Quant Letter No. 45 (2024-04-17): https://www.ml-quant.com/issues/2024-04-17/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Econometrics & Forecasting

## Related

- [TKAN: Temporal Kolmogorov-Arnold Networks](https://www.ml-quant.com/papers/ssrn/4825654/): The article presents Temporal Kolomogorov-Arnold Networks (TKANs), a new neural network design that merges the benefits of Recurrent Neural Networks and Long Short-Term Memory for improved multistep time series forecasting.
- [Robust agents learn causal world models](https://www.ml-quant.com/papers/arxiv/2402.10877/): The research suggests that intelligent agents must learn an approximate causal model to generalize to new domains, impacting fields like transfer learning and causal inference.
- [Regularization for electricity price forecasting](https://www.ml-quant.com/papers/arxiv/2404.03968/): The study reveals that LQ and elastic net penalty functions provide more precise electricity price predictions than other methods, including the popular LASSO. It also confirms that cross-validation is a useful tool for optimizing parameters.
- [A Neural Framework for Generalized Causal Sensitivity Analysis](https://www.ml-quant.com/papers/arxiv/2311.16026/): Causal Sensitivity Analysis: The article introduces NeuralCSA, a new neural framework for analyzing causal sensitivity, capable of handling various models, treatments, and queries, and providing accurate causal query bounds.
- [Supervised autoencoder MLP for financial time series forecasting](https://www.ml-quant.com/papers/arxiv/2404.01866/): The study investigates the use of supervised autoencoders in improving financial forecasting through precise parameter tuning.
- [A Dynamic Regime-Switching Model Using Gated Recurrent Straight-Through Units](https://www.ml-quant.com/papers/ssrn/4810879/): The Gated Recurrent Straightthrough Unit (GRSTU), a new deep learning model, outperforms statistical jump models in identifying regime changes in the S&P500 index, especially with smaller datasets.