# Ensembles of Localised Models for Time Series Forecasting

arXiv: https://arxiv.org/abs/2012.15059 · HTML: https://mlsanity.com/paper/2012.15059/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Dec 2020.

## Built on by (most cited first, 20 of 20)

- [Monash Time Series Forecasting Archive](https://mlsanity.com/paper/2105.06643/) (2021)
- [Forecast Evaluation for Data Scientists: Common Pitfalls and Best Practices](https://mlsanity.com/paper/2203.10716/) (2022)
- [pTSE: A Multi-model Ensemble Method for Probabilistic Time Series Forecasting](https://mlsanity.com/paper/2305.11304/) (2023)
- [Handling Concept Drift in Global Time Series Forecasting](https://mlsanity.com/paper/2304.01512/) (2023)
- [Scalable Probabilistic Forecasting in Retail with Gradient Boosted Trees: A Practitioner's Approach](https://mlsanity.com/paper/2311.00993/) (2023)
- [Your time series is worth a binary image: machine vision assisted deep framework for time series forecasting](https://mlsanity.com/paper/2302.14390/) (2023)
- [On the retraining frequency of global models in retail demand forecasting](https://mlsanity.com/paper/2505.00356/) (2025)
- [SETAR-Tree: A Novel and Accurate Tree Algorithm for Global Time Series Forecasting](https://mlsanity.com/paper/2211.08661/) (2022)
- [Improving Forecasts for Heterogeneous Time Series by "Averaging", with Application to Food Demand Forecast](https://mlsanity.com/paper/2306.07119/) (2023)
- [Time Series Data Augmentation as an Imbalanced Learning Problem](https://mlsanity.com/paper/2404.18537/) (2024)
- [Lag Selection for Univariate Time Series Forecasting using Deep Learning: An Empirical Study](https://mlsanity.com/paper/2405.11237/) (2024)
- [ModelRadar: Aspect-based Forecast Evaluation](https://mlsanity.com/paper/2504.00059/) (2025)
- [Forecasting with Hyper-Trees](https://mlsanity.com/paper/2405.07836/) (2024)
- [Forecasting with Deep Learning: Beyond Average of Average of Average Performance](https://mlsanity.com/paper/2406.16590/) (2024)
- [The cost of ensembling: is it always worth combining?](https://mlsanity.com/paper/2506.04677/) (2025)
- [Selective Time Series Forecasting via Metalearning](https://mlsanity.com/paper/2606.23448/) (2026)
- [Meta-learning and Data Augmentation for Stress Testing Forecasting Models](https://mlsanity.com/paper/2406.17008/) (2024)
- [Designing Time-Series Models With Hypernetworks & Adversarial Portfolios](https://mlsanity.com/paper/2407.20352/) (2024)
- [A Novel Hybrid Approach to Contraceptive Demand Forecasting: Integrating Point Predictions with Probabilistic Distributions](https://mlsanity.com/paper/2502.09685/) (2025)
- [Multipath Adaptive Gated Bottleneck Latent ODE with Raman Data Fusion for Cell Culture Process Forecasting](https://mlsanity.com/paper/2606.26520/) (2026)

## Abstract (arXiv)

With large quantities of data typically available nowadays, forecasting models that are trained across sets of time series, known as Global Forecasting Models (GFM), are regularly outperforming traditional univariate forecasting models that work on isolated series. As GFMs usually share the same set of parameters across all time series, they often have the problem of not being localised enough to a particular series, especially in situations where datasets are heterogeneous. We study how ensembling techniques can be used with generic GFMs and univariate models to solve this issue. Our work systematises and compares relevant current approaches, namely clustering series and training separate submodels per cluster, the so-called ensemble of specialists approach, and building heterogeneous ensembles of global and local models. We fill some gaps in the existing GFM localisation approaches, in particular by incorporating varied clustering techniques such as feature-based clustering, distance-based clustering and random clustering, and generalise them to use different underlying GFM model types. We then propose a new methodology of clustered ensembles where we train multiple GFMs on different clusters of series, obtained by changing the number of clusters and cluster seeds. Using Feed-forward Neural Networks, Recurrent Neural Networks, and Pooled Regression models as the underlying GFMs, in our evaluation on eight publicly available datasets, the proposed models are able to achieve significantly higher accuracy than baseline GFM models and univariate forecasting methods.