A Note on Verification of Prognostic Charts
Abstract
In recent forecasting experiments it has been customary to verify the prognostic charts by computing the correlation coefficients between the observed and redicted pressure changes. The justification has been that correlations between predicted and observed pressures would not correctly reflect skill, since in many cases high correlations would result from persistence.
DOI: https://doi.org/10.3402/tellusa.v9i3.9109 | Journal eISSN: 3035-9554
Language: English
Page range: 313 - 315
Published on: Jan 1, 1957
Published by: Stockholm University Press
In partnership with: Paradigm Publishing Services
© 1957 Sverre Petterssen, published by Stockholm University Press
This work is licensed under the Creative Commons Attribution 4.0 License.