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