Two-Step Meta-Learning for Time-Series Forecasting Ensemble
| Autor: | Vilius Kontrimas, Evaldas Vaiciukynas, Rimantas Butleris, Paulius Danenas |
|---|---|
| Přispěvatelé: | IEEE (Institute of Electrical and Electronics Engineers) |
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning I.5 General Computer Science Rank (linear algebra) Meta learning (computer science) Pooling G.3 Machine Learning (stat.ML) business intelligence Machine Learning (cs.LG) Methodology (stat.ME) meta-learning Statistics - Machine Learning 91B84 Statistics General Materials Science Time series Statistics - Methodology Mathematics M4 competition Series (mathematics) General Engineering Univariate TK1-9971 Random forest univariate time-series model forecasting ensemble Electrical engineering. Electronics. Nuclear engineering Symmetric mean absolute percentage error random forest |
| Zdroj: | IEEE Access, Vol 9, Pp 62687-62696 (2021) |
| ISSN: | 2169-3536 |
| DOI: | 10.1109/access.2021.3074891 |
| Popis: | 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. Accepted to IEEE Access journal in April 22, 2021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|