statsmodels.tools.eval_measures.rmspe¶

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

Root Mean Squared Percentage Error

Parameters:¶

y : array_like¶

The actual value.

y_hat : array_like¶

The predicted value.

axis : int¶

Axis along which the summary statistic is calculated

zeros : float¶

Value to assign to error where y is zero

Returns:¶

rmspe – Root Mean Squared Percentage Error along given axis.

Return type:¶

ndarray or float