Online learning with the Continuous Ranked Probability Score for ensemble forecasting

Quarterly Journal of the Royal Meteorological Society - Tập 143 Số 702 - Trang 521-529 - 2017
Jean Thorey1,2, Vivien Mallet2, P. Baudin2
1CEREA, Joint Research Laboratory ENPC ParisTech—EDF R&D Université Paris‐Est Marne‐La‐Vallée France
2Inria, Paris, France

Tóm tắt

Ensemble forecasting resorts to multiple individual forecasts to produce a discrete probability distribution that represents the uncertainties accurately. Before every forecast, a weighted empirical distribution function is derived from the ensemble, so as to minimize the Continuous Ranked Probability Score (CRPS). We apply online learning techniques that have previously been used for deterministic forecasting and adapt them for the minimization of the CRPS. The proposed method guarantees theoretically that the aggregated forecast competes, in terms of CRPS, against the best weighted empirical distribution function with weights constant in time. This is illustrated on synthetic data. Additionally, our study improves knowledge of the CRPS expectation for model mixtures. We generalize results about the bias of the CRPS computed with ensemble forecasts and propose a new scheme to achieve fair CRPS minimization, without any assumption about the distribution.

Từ khóa


Tài liệu tham khảo

10.1214/08-AOS623

10.1109/TIT.2011.2104610

10.1175/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2

10.1002/qj.456

10.1002/qj.1891

10.1175/WAF966.1

10.1256/qj.04.71

10.1007/b99352

10.1017/CBO9780511546921

10.1111/j.1539-6924.1999.tb00399.x

10.1007/s11749-008-0118-6

10.1007/s10994-012-5314-7

10.1175/1520-0450(1969)008<0985:ASSFPF>2.0.CO;2

10.1002/qj.2270

10.1002/met.45

10.1175/2009MWR3046.1

10.1002/met.1409

10.1002/for.3980090106

10.1146/annurev-statistics-062713-085831

10.1198/016214506000001437

10.1175/MWR2904.1

Good IJ, 1952, Rational decisions, J. R. Stat. Soc., Ser. B (Methodological), 14, 107, 10.1111/j.2517-6161.1952.tb00104.x

10.1256/qj.05.235

10.1175/1520-0434(2000)015<0559:DOTCRP>2.0.CO;2

10.1175/MWR-D-15-0095.1

10.1006/inco.1996.2612

10.1016/j.jcp.2007.02.014

10.1175/MWR2949.1

10.1029/2010JD014259

V Mallet B Mauricette G Stoltz 2007 École Normale Supérieure de Paris Paris

10.1029/2008JD009978

10.1175/1520-0450(1971)010<0155:ANOTRP>2.0.CO;2

10.1007/s10994-014-5474-8

10.1175/MWR2906.1

10.1111/j.1467-9868.2009.00726.x

10.1080/01621459.1971.10482346

Stoltz G, 2010, Agrégation séquentielle de prédicteurs: méthodologie générale et applications à la prévision de la qualité de l'air et à celle de la consommation électrique, J. Soc. Fr. Stat., 151, 66

Vovk V, 2009, Prediction with expert advice for the Brier game, J. Mach. Learn. Res., 10, 2445

10.1175/1520-0450(1968)007<0751:PA>2.0.CO;2

Yitzhaki S, 2012, The Gini Methodology: A Primer on a Statistical Methodology

Thông tin
Thông tin xuất bản

Quarterly Journal of the Royal Meteorological Society

Tập 143 Số 702

521-529

Thông tin tác giả