Standard error of prediction

Standard error of prediction (SEP) is the sample standard deviation of prediction errors after removing their mean error (ME).

Usage

sep(obs, pred, na.rm = TRUE)

Arguments

obs

Numeric observation vector.

pred

Numeric prediction vector paired with obs.

na.rm

Logical; remove incomplete pairs?

Value

One numeric value.

Details

$$\mathrm{SEP}=\sqrt{\frac{1}{n-1}\sum_{i=1}^{n} \left[(obs_i-pred_i)-\frac{1}{n}\sum_{j=1}^{n}(obs_j-pred_j)\right]^2}.$$

SEP has response units, is non-negative, and zero is ideal. Unlike RMSE, it removes constant bias. It returns NA with a warning when fewer than two retained pairs remain. Missing-value handling follows bias().

References

Bellon-Maurel et al. (2010). See rpd().

See also

Other prediction metrics: mape(), mdae(), mpe(), msle(), rae(), rer(), rmsle(), rpd(), rpiq(), rrmse(), smape(), willmott_d()