metric mase

[moghadas76]

Hi,

Are MASE and MAE(in the literature) metric are consistent?

def mase( target: np.ndarray, forecast: np.ndarray, seasonal_error: float, ) -> float: r""" .. math:: mase = mean(|Y - \hat{Y}|) / seasonal\_error See [HA21]_ for more details. """ return np.mean(np.abs(target - forecast)) / seasonal_error

My implementation for MAE:

def mae( target: np.ndarray, forecast: np.ndarray, seasonal_error: float, ) -> float: r""" .. math:: mase = mean(|Y - \hat{Y}|) See [HA21]_ for more details. """ return np.mean(np.abs(target - forecast))