brier {verification}R Documentation

Brier Score

Description

Calculates verification statistics for probabilistic forecasts of binary events.

Usage

    brier(obs, pred, baseline, thresholds = seq(0,1,0.1), bins = TRUE, ... )
       

Arguments

obs Vector of binary observations
pred Vector of probablistic predictions [0,1]
baseline Vector of climatological (no - skill) forecasts. If this is null, a sample climatology will be calculated.
thresholds Values used to bin the forecasts. By default the bins are {[0,0.1), [0.1, 0.2), ....} .
bins If TRUE, thresholds define bins into which the probablistic forecasts are entered and assigned the midpoint as a forecast. Otherwise, each unique forecast is considered as a seperate forecast. For example, set bins to FALSE when dealing with a finite number of probabilities generated by an ensemble forecast.
... Optional arguments

Value

baseline.tf Logical indicator of whether climatology was provided.
bs Brier score
bs.baseline Brier Score for climatology
ss Skill score
bs.reliability Reliability portion of Brier score.
bs.resolution Resolution component of Brier score.
bs.uncert Uncertainty component of Brier score.
y.i Forecast bins – described as the center value of the bins.
obar.i Observation bins – described as the center value of the bins.
prob.y Proportion of time using each forecast
obar Forecast based on climatology or average sample observations.
check Reliability - resolution + uncertainty should equal brier score.

Note

This function is used within verify.

Author(s)

Matt Pocernich <pocernic@rap.ucar.edu>

References

Wilks, D. S. (1995) Statistical Methods in the Atmospheric Sciences Chapter 7, San Diego: Academic Press.

Examples


#  probabilistic/ binary example

pred<- runif(100)
obs<- round(runif(100))
brier(obs, pred)


[Package verification version 1.29 Index]