👉 In the context of statistics and machine learning, "brier" refers to a measure of the accuracy or bias of an estimator, particularly in regression analysis. The Brier score is defined as:
Brier Score = 2
log(1 - P(y = y
) / (1 - P(y = y
)))
Where:
- P(y) is the probability of observing y if x were not a predictor,
- y
and y are the values predicted by the estimator, and