mlsanityData to 14 Oct 2026

Forecast combinations: an over 50-year review

arXiv 2205.04216 ↗First posted 9 May 2022stat.ME stat.AP stat.COcited by 34

Not on the map. Why?

Abstract (arXiv)

Forecast combinations have flourished remarkably in the forecasting community and, in recent years, have become part of the mainstream of forecasting research and activities. Combining multiple forecasts produced from single (target) series is now widely used to improve accuracy through the integration of information gleaned from different sources, thereby mitigating the risk of identifying a single "best" forecast. Combination schemes have evolved from simple combination methods without estimation, to sophisticated methods involving time-varying weights, nonlinear combinations, correlations among components, and cross-learning. They include combining point forecasts and combining probabilistic forecasts. This paper provides an up-to-date review of the extensive literature on forecast combinations, together with reference to available open-source software implementations. We discuss the potential and limitations of various methods and highlight how these ideas have developed over time. Some important issues concerning the utility of forecast combinations are also surveyed. Finally, we conclude with current research gaps and potential insights for future research.

Key facts. Cited by 34 papers in the map (data through 14 Oct 2026). Builds on 25 of its 285 references that are in the map. First posted on arXiv 9 May 2022.

Builds on

25 of 285 references.

References · 25

  1. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 753
  2. 2021Ensemble deep learning: A reviewcited by 284
  3. 2019catch22: CAnonical Time-series CHaracteristicscited by 105
  4. 2018Neural networks for post-processing ensemble weather forecastscited by 95
  5. 2020Deep Learning for Post-Processing Ensemble Weather Forecastscited by 52
  6. 2019GRATIS: GeneRAting TIme Series with diverse and controllable characteristicscited by 42
  7. 2017Using stacking to average Bayesian predictive distributionscited by 39
  8. 2020Forecasting: theory and practicecited by 36
  9. 2019Forecasting with time series imagingcited by 22
  10. 2021Flexible Model Aggregation for Quantile Regressioncited by 22
  11. 2021CRPS Learningcited by 21
  12. 2016Automatic time-series phenotyping using massive feature extractioncited by 12
13 more
  1. 2016Combining predictive distributions for statistical post-processing of ensemble forecastscited by 9
  2. 2020Forecast with Forecasts: Diversity Matterscited by 8
  3. 2016Dynamic Bayesian Predictive Synthesis in Time Series Forecastingcited by 7
  4. 2015Bayesian Nonparametric Calibration and Combination of Predictive Distributionscited by 6
  5. 2017Model Averaging and its Use in Economicscited by 5
  6. 2017Forest-based methods and ensemble model output statistics for rainfall ensemble forecastingcited by 5
  7. 2021An Empirical Evaluation of Time-Series Feature Setscited by 5
  8. 2017Multivariate Bayesian Predictive Synthesis in Macroeconomic Forecastingcited by 4
  9. 2022Comparing trained and untrained probabilistic ensemble forecasts of COVID-19 cases and deaths in the United Statescited by 4
  10. 2013Probabilistic wind speed forecasting using Bayesian model averaging with truncated normal componentscited by 2
  11. 2020Optimal probabilistic forecasts: When do they work?cited by 2
  12. 2015On the Forecast Combination Puzzlecited by 1
  13. 2020For2For: Learning to forecast from forecastscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 7; 2024: 11; 2025: 9; 2026: 6.

34 papers.

Sort:

  1. 2023A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detectioncited by 145
  2. 2023AutoGluon-TimeSeries: AutoML for Probabilistic Time Series Forecastingcited by 36
  3. 2022A review of machine learning concepts and methods for addressing challenges in probabilistic hydrological post-processing and forecastingcited by 12
  4. 2024On-line conformalized neural networks ensembles for probabilistic forecasting of day-ahead electricity pricescited by 5
  5. 2023Ensemble learning for blending gridded satellite and gauge-measured precipitation datacited by 3
  6. 2023Combining predictive distributions of electricity prices: Does minimizing the CRPS lead to optimal decisions in day-ahead bidding?cited by 3
  7. 2024Ensemble learning for uncertainty estimation with application to the correction of satellite precipitation productscited by 3
  8. 2024Stacking for Probabilistic Short-term Load Forecastingcited by 2
  9. 2025Comparative analysis and evaluation of ageing forecasting methods for semiconductor devices in online health monitoringcited by 2
  10. 2025Test-Time Efficient Pretrained Model Portfolios for Time Series Forecastingcited by 2
  11. 2024Model Averaging and Double Machine Learningcited by 1
  12. 2024Maximally Forward-Looking Core Inflationcited by 1
22 more
  1. 2024Time-Series Forecasting in Smart Manufacturing Systems: An Experimental Evaluation of the State-of-the-art Algorithmscited by 1
  2. 2025The cost of ensembling: is it always worth combining?cited by 1
  3. 2023Optimizing accuracy and diversity: a multi-task approach to forecast combinationscited by 0
  4. 2023Infinite forecast combinations based on Dirichlet processcited by 0
  5. 2023Enhancing Algorithm Performance Understanding through tsMorph: Generating Semi-Synthetic Time Series for Robust Forecasting Evaluationcited by 0
  6. 2024Operational Collective Intelligence of Humans and Machinescited by 0
  7. 2024Movement Prediction-Adjusted Naive Forecast: Is the Naive Baseline Unbeatable in Financial Time Series Forecasting?cited by 0
  8. 2024Enhancing reliability in prediction intervals using point forecasters: Heteroscedastic Quantile Regression and Width-Adaptive Conformal Inferencecited by 0
  9. 2024Combinations of distributional regression algorithms with application in uncertainty estimation of corrected satellite precipitation productscited by 0
  10. 2024Efficient pooling of predictions via kernel embeddingscited by 0
  11. 2025High-dimensional point forecast combinations for emergency department demandcited by 0
  12. 2025A Novel Hybrid Approach to Contraceptive Demand Forecasting: Integrating Point Predictions with Probabilistic Distributionscited by 0
  13. 2025Combining Forecasts using Meta-Learning: A Comparative Study for Complex Seasonalitycited by 0
  14. 2025Probabilistic Prediction Markets with Intermittent Contributionscited by 0
  15. 2025Data-driven Calibration Sample Selection and Forecast Combination in Electricity Price Forecasting: An Application of the ARHNN Methodcited by 0
  16. 2025Multi-layer Stack Ensembles for Time Series Forecastingcited by 0
  17. 2026Deployment-Side Adaptiveness in Multi-Horizon Volatility Forecastingcited by 0
  18. 2026Rollcast: Proper-Score Gated Rolling Anchors for Adaptive Probabilistic Time-Series Forecastingcited by 0
  19. 2026Ensemble Complexity in Photovoltaic Forecastingcited by 0
  20. 2026Stealing profits: Spread-based temporal hierarchy forecasting for day-ahead electricity marketscited by 0
  21. 2026Target alignment, dilution and forecast selection when cross-sectional forecasts share a common targetcited by 0
  22. 2026Downside-Controlled Online Forecast Combination under Delayed and Revised Outcomescited by 0