TY - JOUR
T1 - The uncertainty estimation of feature-based forecast combinations
AU - Wang, Xiaoqian
AU - Kang, Yanfei
AU - Petropoulos, Fotios
AU - Li, Feng
N1 - Funding Information:
Yanfei Kang is supported by the National Key Research and Development Program (No. 2019YFB1404600) and the National Natural Science Foundation of China (No.11701022). Feng Li is supported by the National Natural Science Foundation of China (No. 11501587) and the Beijing Universities Advanced Disciplines Initiative (No. GJJ2019163). Petropoulos' work was completed during his visit at the Beihang University in April-May 2019. This research was supported by the high-performance computing (HPC) resources at Beihang University.
Publisher Copyright:
© Operational Research Society 2021.
PY - 2022
Y1 - 2022
N2 - Forecasting is an indispensable element of operational research (OR) and an important aid to planning. The accurate estimation of the forecast uncertainty facilitates several operations management activities, predominantly in supporting decisions in inventory and supply chain management and effectively setting safety stocks. In this paper, we introduce a feature-based framework, which links the relationship between time series features and the interval forecasting performance into providing reliable interval forecasts. We propose an optimal threshold ratio searching algorithm and a new weight determination mechanism for selecting an appropriate subset of models and assigning combination weights for each time series tailored to the observed features. We evaluate our approach using a large set of time series from the M4 competition. Our experiments show that our approach significantly outperforms a wide range of benchmark models, both in terms of point forecasts as well as prediction intervals.
AB - Forecasting is an indispensable element of operational research (OR) and an important aid to planning. The accurate estimation of the forecast uncertainty facilitates several operations management activities, predominantly in supporting decisions in inventory and supply chain management and effectively setting safety stocks. In this paper, we introduce a feature-based framework, which links the relationship between time series features and the interval forecasting performance into providing reliable interval forecasts. We propose an optimal threshold ratio searching algorithm and a new weight determination mechanism for selecting an appropriate subset of models and assigning combination weights for each time series tailored to the observed features. We evaluate our approach using a large set of time series from the M4 competition. Our experiments show that our approach significantly outperforms a wide range of benchmark models, both in terms of point forecasts as well as prediction intervals.
KW - forecast combination
KW - Forecasting
KW - prediction intervals
KW - time series features
KW - uncertainty estimation
UR - http://www.scopus.com/inward/record.url?scp=85130083265&partnerID=8YFLogxK
U2 - 10.1080/01605682.2021.1880297
DO - 10.1080/01605682.2021.1880297
M3 - Article
AN - SCOPUS:85130083265
SN - 0160-5682
VL - 73
SP - 979
EP - 993
JO - Journal of the Operational Research Society
JF - Journal of the Operational Research Society
IS - 5
ER -