Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory

Abstract In this study, we modeled the multi‐model blending process using random variables and explicitly derived the distribution of the blended forecast error under the assumption of normally distributed errors. Utilizing this error distribution, we regained the scalar version of the Best Linear U...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu Wang, Yong Cao, Yue Shen, Ruixia Zhao, Xiaoqing Zeng, Li Yao, Kan Dai
Format: Article
Language:English
Published: Wiley 2025-02-01
Series:Geophysical Research Letters
Subjects:
Online Access:https://doi.org/10.1029/2024GL111622
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850032923651080192
author Yu Wang
Yong Cao
Yue Shen
Ruixia Zhao
Xiaoqing Zeng
Li Yao
Kan Dai
author_facet Yu Wang
Yong Cao
Yue Shen
Ruixia Zhao
Xiaoqing Zeng
Li Yao
Kan Dai
author_sort Yu Wang
collection DOAJ
description Abstract In this study, we modeled the multi‐model blending process using random variables and explicitly derived the distribution of the blended forecast error under the assumption of normally distributed errors. Utilizing this error distribution, we regained the scalar version of the Best Linear Unbiased Estimator. Notably, the model yielded negative weights, which contradict traditional assumptions that weights should be positive. To address this, we modified the algorithm to set negative weights to zero and compared the performance with the original algorithm. Our findings indicate that setting negative weights to zero results in a slight improvement in blending performance compared to using negative weights directly. This improvement may be attributed to modeling the actual forecast error as a normally distributed unified parameter and applying a consistent correlation coefficient across the annual data set.
format Article
id doaj-art-e474b630aac04df683fa95aed8bb0f3c
institution DOAJ
issn 0094-8276
1944-8007
language English
publishDate 2025-02-01
publisher Wiley
record_format Article
series Geophysical Research Letters
spelling doaj-art-e474b630aac04df683fa95aed8bb0f3c2025-08-20T02:58:26ZengWileyGeophysical Research Letters0094-82761944-80072025-02-01524n/an/a10.1029/2024GL111622Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables TheoryYu Wang0Yong Cao1Yue Shen2Ruixia Zhao3Xiaoqing Zeng4Li Yao5Kan Dai6National Meteorological Center of China Meteorological Administration Beijing PR. ChinaNational Meteorological Center of China Meteorological Administration Beijing PR. ChinaChina Meteorological Administration Training Center Beijing PR. ChinaNational Meteorological Center of China Meteorological Administration Beijing PR. ChinaNational Meteorological Center of China Meteorological Administration Beijing PR. ChinaNational Meteorological Center of China Meteorological Administration Beijing PR. ChinaNational Meteorological Center of China Meteorological Administration Beijing PR. ChinaAbstract In this study, we modeled the multi‐model blending process using random variables and explicitly derived the distribution of the blended forecast error under the assumption of normally distributed errors. Utilizing this error distribution, we regained the scalar version of the Best Linear Unbiased Estimator. Notably, the model yielded negative weights, which contradict traditional assumptions that weights should be positive. To address this, we modified the algorithm to set negative weights to zero and compared the performance with the original algorithm. Our findings indicate that setting negative weights to zero results in a slight improvement in blending performance compared to using negative weights directly. This improvement may be attributed to modeling the actual forecast error as a normally distributed unified parameter and applying a consistent correlation coefficient across the annual data set.https://doi.org/10.1029/2024GL111622optimal blendingmulti‐model blendingrandom variablesstochastic processesuncertainty quantification
spellingShingle Yu Wang
Yong Cao
Yue Shen
Ruixia Zhao
Xiaoqing Zeng
Li Yao
Kan Dai
Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory
Geophysical Research Letters
optimal blending
multi‐model blending
random variables
stochastic processes
uncertainty quantification
title Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory
title_full Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory
title_fullStr Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory
title_full_unstemmed Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory
title_short Explanation and Optimizing Multi‐Model Blending Algorithm Using Random Variables Theory
title_sort explanation and optimizing multi model blending algorithm using random variables theory
topic optimal blending
multi‐model blending
random variables
stochastic processes
uncertainty quantification
url https://doi.org/10.1029/2024GL111622
work_keys_str_mv AT yuwang explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory
AT yongcao explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory
AT yueshen explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory
AT ruixiazhao explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory
AT xiaoqingzeng explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory
AT liyao explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory
AT kandai explanationandoptimizingmultimodelblendingalgorithmusingrandomvariablestheory