Forecasting skill of model averages

C Larrabee Winter, Doug Nychka

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

Given a collection of science-based computational models that all estimate states of the same environmental system, we compare the forecast skill of the average of the collection to the skills of the individual members. We illustrate our results through an analysis of regional climate model data and give general criteria for the average to perform more or less skillfully than the most skillful individual model, the "best" model. The average will only be more skillful than the best model if the individual models in the collection produce very different forecasts; if the individual forecasts generally agree, the average will not be as skillful as the best model.

Original languageEnglish (US)
Pages (from-to)633-638
Number of pages6
JournalStochastic Environmental Research and Risk Assessment
Volume24
Issue number5
DOIs
StatePublished - 2010

Fingerprint

Climate models
regional climate
climate modeling
forecast
analysis
science

Keywords

  • Environmental models
  • Mean-square error
  • Model average
  • Model comparison
  • Stochastic processes
  • Uncertainty

ASJC Scopus subject areas

  • Environmental Engineering
  • Environmental Science(all)
  • Environmental Chemistry
  • Water Science and Technology
  • Safety, Risk, Reliability and Quality

Cite this

Forecasting skill of model averages. / Winter, C Larrabee; Nychka, Doug.

In: Stochastic Environmental Research and Risk Assessment, Vol. 24, No. 5, 2010, p. 633-638.

Research output: Contribution to journalArticle

@article{d2b7dae1900b4933b23bd80d230ebefb,
title = "Forecasting skill of model averages",
abstract = "Given a collection of science-based computational models that all estimate states of the same environmental system, we compare the forecast skill of the average of the collection to the skills of the individual members. We illustrate our results through an analysis of regional climate model data and give general criteria for the average to perform more or less skillfully than the most skillful individual model, the {"}best{"} model. The average will only be more skillful than the best model if the individual models in the collection produce very different forecasts; if the individual forecasts generally agree, the average will not be as skillful as the best model.",
keywords = "Environmental models, Mean-square error, Model average, Model comparison, Stochastic processes, Uncertainty",
author = "Winter, {C Larrabee} and Doug Nychka",
year = "2010",
doi = "10.1007/s00477-009-0350-y",
language = "English (US)",
volume = "24",
pages = "633--638",
journal = "Stochastic Environmental Research and Risk Assessment",
issn = "1436-3240",
publisher = "Springer New York",
number = "5",

}

TY - JOUR

T1 - Forecasting skill of model averages

AU - Winter, C Larrabee

AU - Nychka, Doug

PY - 2010

Y1 - 2010

N2 - Given a collection of science-based computational models that all estimate states of the same environmental system, we compare the forecast skill of the average of the collection to the skills of the individual members. We illustrate our results through an analysis of regional climate model data and give general criteria for the average to perform more or less skillfully than the most skillful individual model, the "best" model. The average will only be more skillful than the best model if the individual models in the collection produce very different forecasts; if the individual forecasts generally agree, the average will not be as skillful as the best model.

AB - Given a collection of science-based computational models that all estimate states of the same environmental system, we compare the forecast skill of the average of the collection to the skills of the individual members. We illustrate our results through an analysis of regional climate model data and give general criteria for the average to perform more or less skillfully than the most skillful individual model, the "best" model. The average will only be more skillful than the best model if the individual models in the collection produce very different forecasts; if the individual forecasts generally agree, the average will not be as skillful as the best model.

KW - Environmental models

KW - Mean-square error

KW - Model average

KW - Model comparison

KW - Stochastic processes

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=77955054926&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955054926&partnerID=8YFLogxK

U2 - 10.1007/s00477-009-0350-y

DO - 10.1007/s00477-009-0350-y

M3 - Article

VL - 24

SP - 633

EP - 638

JO - Stochastic Environmental Research and Risk Assessment

JF - Stochastic Environmental Research and Risk Assessment

SN - 1436-3240

IS - 5

ER -