Resolve "release v2.3.0"

Closes #440 (closed)

mlair/helpers/statistics.py

@@ -214,6 +214,11 @@ def mean_absolute_error(a, b, dim=None):
     return np.abs(a - b).mean(dim)
 
 
+def mean_error(a, b, dim=None):
+    """Calculate mean error where a is forecast and b the reference (e.g. observation)."""
+    return a.mean(dim) - b.mean(dim)
+
+
 def index_of_agreement(a, b, dim=None):
     """Calculate index of agreement (IOA) where a is the forecast and b the reference (e.g. observation)."""
     num = (np.square(b - a)).sum(dim)
@@ -235,7 +240,7 @@ def modified_normalized_mean_bias(a, b, dim=None):
 
 
 def calculate_error_metrics(a, b, dim):
-    """Calculate MSE, RMSE, MAE, IOA, and MNMB. Additionally, return number of used values for calculation.
+    """Calculate MSE, ME, RMSE, MAE, IOA, and MNMB. Additionally, return number of used values for calculation.
 
     :param a: forecast data to calculate metrics for
     :param b: reference (e.g. observation)
@@ -244,22 +249,25 @@ def calculate_error_metrics(a, b, dim):
     :returns: dict with results for all metrics indicated by lowercase metric short name
     """
     mse = mean_squared_error(a, b, dim)
+    me = mean_error(a, b, dim)
     rmse = np.sqrt(mse)
     mae = mean_absolute_error(a, b, dim)
     ioa = index_of_agreement(a, b, dim)
     mnmb = modified_normalized_mean_bias(a, b, dim)
     n = (a - b).notnull().sum(dim)
-    results = {"mse": mse, "rmse": rmse, "mae": mae, "ioa": ioa, "mnmb": mnmb, "n": n}
+    results = {"mse": mse, "me": me, "rmse": rmse, "mae": mae, "ioa": ioa, "mnmb": mnmb, "n": n}
     return {k: squeeze_coords(v) for k, v in results.items()}
 
 
 def get_error_metrics_units(base_unit):
-    return {"mse": f"{base_unit}$^2$", "rmse": base_unit, "mae": base_unit, "ioa": None, "mnmb": None, "n": None}
+    return {"mse": f"{base_unit}$^2$", "me": base_unit, "rmse": base_unit, "mae": base_unit, "ioa": None, "mnmb": None,
+            "n": None}
 
 
 def get_error_metrics_long_name():
-    return {"mse": "mean squared error", "rmse": "root mean squared error", "mae": "mean absolute error",
-            "ioa": "index of agreement", "mnmb": "modified normalized mean bias", "n": "count"}
+    return {"mse": "mean squared error", "me": "mean error", "rmse": "root mean squared error",
+            "mae": "mean absolute error", "ioa": "index of agreement", "mnmb": "modified normalized mean bias",
+            "n": "count"}
 
 
 def mann_whitney_u_test(data: pd.DataFrame, reference_col_name: str, **kwargs):