Symmetric mean absolute percentage error

Symmetric mean absolute percentage error (sMAPE) scales absolute error by absolute observation and prediction sizes.

Usage

smape(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

$$\mathrm{sMAPE}=\frac{100}{n}\sum_{i=1}^n \frac{2|obs_i-pred_i|}{|obs_i|+|pred_i|}.$$

sMAPE ranges from zero to 200 percent; zero is ideal. A pair of zeros contributes zero. Missing-value handling follows bias().

References

Hyndman and Koehler (2006). See mdae().

See also

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