EconBase
← All papers

CRPS Learning

Jonathan Berrisch, Florian Ziel

arXiv 1 Feb 2021 · Statistics — Machine Learning

arXiv:2102.00968 · PDF · Extracted main text

Abstract

Combination and aggregation techniques can significantly improve forecast accuracy. This also holds for probabilistic forecasting methods where predictive distributions are combined. There are several time-varying and adaptive weighting schemes such as Bayesian model averaging (BMA). However, the quality of different forecasts may vary not only over time but also within the distribution. For example, some distribution forecasts may be more accurate in the center of the distributions, while others are better at predicting the tails. Therefore, we introduce a new weighting method that considers the differences in performance over time and within the distribution. We discuss pointwise combination algorithms based on aggregation across quantiles that optimize with respect to the continuous ranked probability score (CRPS). After analyzing the theoretical properties of pointwise CRPS learning, we discuss B- and P-Spline-based estimation techniques for batch and online learning, based on quantile regression and prediction with expert advice. We prove that the proposed fully adaptive Bernstein online aggregation (BOA) method for pointwise CRPS online learning has optimal convergence properties. They are confirmed in simulations and a probabilistic forecasting study for European emission allowance (EUA) prices.

Citation extraction

77
references
131
in-text mentions
77
distinct cited
3
self-citations
14,607
main-text words

appendix boundary found by appendix_titled_section at “Appendix” · 95% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Wintenberger, O (2017) Optimal learning with Bernstein online aggregation0.97112492%
2Gaillard, P., Stoltz, G., & Van Erven, T (2014) A second-order bound with excess losses0.92843100%
3Cesa-Bianchi, N., & Lugosi, G (2006) Prediction, learning, and games0.87482100%
4Thorey, J., Chaussin, C., & Mallet, V (2018) Ensemble forecast of photovoltaic power with online CRPS learning0.84333100%
5Gneiting, T (2011) Making and evaluating point forecasts0.73732100%
6Raftery, A. E., Gneiting, T., Balabdaoui, F., & Polakowski, M (2005) Using Bayesian Model Averaging to Calibrate Forecast Ensembles0.73732100%
7Zamo, M., Bel, L., & Mestre, O (2021) Sequential aggregation of probabilistic forecasts—Application to wind speed ensemble forecasts0.73732100%
8Gaillard, P., & Wintenberger, O (2018) Efficient online algorithms for fast-rate regret bounds under sparsity0.6936250%
9Aastveit, K. A., Gerdrup, K. R., Jore, A. S., & Thorsrud, L. A (2014) Nowcasting GDP in real time: A density combination approach0.64422100%
10Aastveit, K. A., Mitchell, J., Ravazzolo, F., & van Dijk, H. K (2019) The Evolution of Forecast Density Combinations in Economics0.64422100%

Showing the top 10 of 77 scored citations.