verification: brier – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

brier

Brier Score

Calculates verification statistics for probabilistic forecasts of binary events.

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

`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

`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.

This function is used within verify.

Matt Pocernich

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

#  probabilistic/ binary example

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

Weather Forecast Verification Utilities

v1.42

GPL (>= 2)

Authors

NCAR - Research Applications Laboratory

Initial release

2015-07-10