GaussCrps {SpecsVerification} | R Documentation |
Calculate the Continuous Ranked Probability Score (CRPS) for forecasts issued as Normal distributions
GaussCrps(mean, sd, obs)
mean |
A vector of length N. The forecast means. |
sd |
A vector of length N. The forecast standard deviations. |
obs |
A numeric vector of length N of real-valued verifying observations |
numeric vector of length N with the CRPS values
Gneiting et al (2005). Calibrated Probabilistic Forecasting Using Ensemble Model Output Statistics and Minimum CRPS Estimation. Mon. Wea. Rev. doi: 10.1175/MWR2904.1
EnsCrps, DressCrps, ScoreDiff, SkillScore
data(eurotempforecast)
mean <- rowMeans(ens)
sd <- apply(ens, 1, sd)
mean(GaussCrps(mean, sd, obs))