statsmodels.tools.eval_measures.rmspe¶

statsmodels.tools.eval_measures.rmspe(y, y_hat, axis=0, zeros=nan)[source]¶

Root Mean Squared Percentage Error

Parameters:¶

yarray_like

The actual value.

y_hatarray_like

The predicted value.

axisint

Axis along which the summary statistic is calculated

zerosfloat

Value to assign to error where y is zero

Returns:¶

rmspendarray or float

Root Mean Squared Percentage Error along given axis.