A Decomposition of the Correlation Coefficient and its Use in Analyzing Forecasting Skill

作者: Thomas R. Stewart

DOI: 10.1175/1520-0434(1990)005<0661:ADOTCC>2.0.CO;2

关键词: Decomposition method (constraint satisfaction)Decomposition (computer science)Consistency (statistics)MathematicsCorrelationCorrelation coefficientStatisticsCognitive systemsHuman judgment

摘要: Abstract Estimates of several components forecasting skill can be obtained by combining a skill-score decomposition developed Allan Murphy with techniques for decomposing correlation coefficients that have been employed in research on human judgment. The the coefficient requires knowledge information or “cues” used forecaster. When cues are known, it is possible to estimate effects uncertainty and forecaster's consistency use cues.

参考文章(0)