Brier score for binary classification problems defined as $$ \frac{1}{n} \sum_{i=1}^n (I_i - p_i)^2. $$ \(I_{i}\) is 1 if observation \(i\) belongs to the positive class, and 0 otherwise.

Note that this (more common) definition of the Brier score is equivalent to the original definition of the multi-class Brier score (see mbrier()) divided by 2.

bbrier(truth, prob, positive, ...)



:: factor()
True (observed) labels. Must have the exactly same two levels and the same length as response.


:: numeric()
Predicted probability for positive class. Must have exactly same length as truth.


:: character(1)
Name of the positive class.


:: any
Additional arguments. Currently ignored.


Performance value as numeric(1).

Meta Information

  • Type: "binary"

  • Range: \([0, 1]\)

  • Minimize: TRUE

  • Required prediction: prob


Brier GW (1950). “Verification of forecasts expressed in terms of probability.” Monthly Weather Review, 78(1), 1--3. doi: 10.1175/1520-0493(1950)078<0001:vofeit>;2 .

See also

Other Binary Classification Measures: auc(), dor(), fbeta(), fdr(), fnr(), fn(), fomr(), fpr(), fp(), mcc(), npv(), ppv(), tnr(), tn(), tpr(), tp()


set.seed(1) lvls = c("a", "b") truth = factor(sample(lvls, 10, replace = TRUE), levels = lvls) prob = runif(10) bbrier(truth, prob, positive = "a")
#> [1] 0.2812546