# Two-Step Meta-Learning for Time-Series Forecasting Ensemble

arXiv: https://arxiv.org/abs/2011.10545 · HTML: https://mlsanity.com/paper/2011.10545/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Nov 2020.

## Built on by (most cited first, 3 of 3)

- [Short-term Renewable Energy Forecasting in Greece using Prophet Decomposition and Tree-based Ensembles](https://mlsanity.com/paper/2107.03825/) (2021)
- [Forecasting Early with Meta Learning](https://mlsanity.com/paper/2307.09796/) (2023)
- [Attention-Based Ensemble Pooling for Time Series Forecasting](https://mlsanity.com/paper/2310.16231/) (2023)

## Abstract (arXiv)

Amounts of historical data collected increase and business intelligence applicability with automatic forecasting of time series are in high demand. While no single time series modeling method is universal to all types of dynamics, forecasting using an ensemble of several methods is often seen as a compromise. Instead of fixing ensemble diversity and size, we propose to predict these aspects adaptively using meta-learning. Meta-learning here considers two separate random forest regression models, built on 390 time-series features, to rank 22 univariate forecasting methods and recommend ensemble size. The forecasting ensemble is consequently formed from methods ranked as the best, and forecasts are pooled using either simple or weighted average (with a weight corresponding to reciprocal rank). The proposed approach was tested on 12561 micro-economic time-series (expanded to 38633 for various forecasting horizons) of M4 competition where meta-learning outperformed Theta and Comb benchmarks by relative forecasting errors for all data types and horizons. Best overall results were achieved by weighted pooling with a symmetric mean absolute percentage error of 9.21% versus 11.05% obtained using the Theta method.