`rmse` – Root mean squared error – `√MSE`.#

Back to point_metrics axis | Back to L5 | Browse all options

Operational op under axis point_metrics, sub-layer L5_A_metric_specification, layer l5. Standalone callable: mf.functions.rmse.

Function signature#

mf.functions.rmse(
    y_true: np.ndarray | pd.Series,
    y_pred: np.ndarray | pd.Series,
) -> float

Parameters#

name	type	default	constraint	description
`y_true`	`np.ndarray	pd.Series`	—	—
`y_pred`	`np.ndarray	pd.Series`	—	—

Returns#

float — scalar result.

Behavior#

Point-forecast metric rmse. Same ranking as MSE but expressed in target units (rather than squared target units). Standard reporting metric in macro / finance papers; pairs naturally with confidence-band charts since RMSE has the same units as the prediction interval.

When to use

Reporting forecast accuracy in target units.

When NOT to use

Heavy-tailed errors – inherits MSE’s outlier sensitivity.

In recipe context#

Set params.point_metrics = "rmse" in the relevant layer to activate this op within a recipe:

# Layer L5 recipe fragment
params:
  point_metrics: rmse

References#

macroforecast design Part 3, L5: ‘evaluation = (metric × benchmark × aggregation × decomposition × ranking).’
Diebold (2017) ‘Forecasting in Economics, Business, Finance and Beyond’, University of Pennsylvania (free online). https://www.sas.upenn.edu/~fdiebold/Textbooks.html

rmse – Root mean squared error – √MSE.#