A function used to quantify the difference between observed data and predicted values according to a model. Minimization of loss functions is a way to estimate the parameters of the model.
- The (Root) Mean Squared Error, rms, used in "ordinary" regression or ordinary least-squares (OLS)
- The Mean Absolute Error, mae, frequently used in forecasting
- "Hinge" losses, or linear losses where over- and underpredictions are weighted differently, for quantile-regression
- (Proper) scoring-rules, used to compare predictive densities to actuals