Ah. Murphy, THE COEFFICIENTS OF CORRELATION AND DETERMINATION AS MEASURES OF PERFORMANCE IN FORECAST VERIFICATION, Weather and forecasting, 10(4), 1995, pp. 681-688
This paper is concerned with the use of the coefficient of correlation
(CoC) and the coefficient of determination (CoD) as performance measu
res in forecast verification. Aspects of forecasting performance that
are measured-and not measured (i.e., ignored)-by these coefficients ar
e identified. Decompositions of familiar quadratic measures of accurac
y and skill are used to explore differences between these quadratic me
asures and the coefficients of correlation and determination. A linear
regression model, in which forecasts are regressed on observations, i
s introduced to provide insight into the interpretations of the CoC an
d the CoD in this context. Issues related to the use of these coeffici
ents as verification measures are discussed, including the deficiencie
s inherent in one-dimensional measures of overall performance, the pro
s and cons of quadratic measures of accuracy and skill vis-a-vis the c
oefficients of correlation and determination, and the relative merits
of the CoC and the CoD. These coefficients by themselves do not provid
e an adequate basis for drawing firm conclusions regarding absolute or
relative forecasting performance.