Generic functions for calculating the Continuous Ranked Probability Score and the Logarithmic Score of R objects.
scoringRules
provides default methods
(crps.numeric
, logs.numeric
) to calculate scores of forecasts
that are members of families of parametric distributions.
crps(y, ...)logs(y, ...)
Returns a vector of scores. One for each forecast-observation pair.
an object for which the score is to be calculated
further arguments passed to or from other methods
The mean logarithmic score corresponds to the negative of the
log-likelihood logLik
.
General background and further references on scoring rules:
Gneiting, T. and A.E. Raftery (2007): `Strictly proper scoring rules, prediction and estimation', Journal of the American Statistical Association 102, 359-378. tools:::Rd_expr_doi("10.1198/016214506000001437")
Gneiting, T. and M. Katzfuss (2014): `Probabilistic forecasting', Annual Review of Statistics and Its Application 1, 125-151. tools:::Rd_expr_doi("10.1146/annurev-statistics-062713-085831")
crps.numeric
, logs.numeric