[ENH] Add test for global forecasting

sktime/forecasting/compose/tests/test_reduce_global.py

@@ -2,15 +2,24 @@
 """Test extraction of features across (shifted) windows."""
 __author__ = ["danbartl"]
 
+import random
+
+import numpy as np
 import pandas as pd
 import pytest
-from sklearn.ensemble import RandomForestRegressor
+from sklearn.ensemble import (
+    GradientBoostingRegressor,
+    HistGradientBoostingRegressor,
+    RandomForestRegressor,
+)
 from sklearn.pipeline import make_pipeline
 
 from sktime.datasets import load_airline
 from sktime.datatypes import get_examples
+from sktime.forecasting.base import ForecastingHorizon
 from sktime.forecasting.compose import make_reduction
 from sktime.forecasting.model_selection import temporal_train_test_split
+from sktime.performance_metrics.forecasting import mean_absolute_percentage_error
 from sktime.transformations.series.summarize import WindowSummarizer
 
 # Load data that will be the basis of tests
@@ -175,3 +184,54 @@ def test_list_reduction(y, index_names):
     forecaster2.fit(y, fh=[1, 2])
     y_pred = forecaster2.predict(fh=[1, 2, 12])
     check_eval(y_pred.index.names, index_names)
+
+
+def test_equality_transfo_nontranso():
+    """Test that recursive reducers return same results for global / local forecasts."""
+    ensemblealgos = [
+        RandomForestRegressor(),
+        GradientBoostingRegressor(),
+        HistGradientBoostingRegressor(),
+    ]
+
+    y = load_airline()
+    y_train, y_test = temporal_train_test_split(y, test_size=30)
+    fh = ForecastingHorizon(y_test.index, is_relative=False)
+
+    lag_vec = [i for i in range(1, 13)]
+    lag_vec.reverse()
+    kwargs = {
+        "lag_feature": {
+            "lag": lag_vec,
+            # "mean": [[1, 52], [3, 52]],
+            # "std": [[1, 52]],
+        }
+    }
+
+    for algo in ensemblealgos:
+        for _i in range(1, 5):
+            random_int = random.randint(1, 1000)
+            regressor = algo
+            regressor.random_state = random_int
+            forecaster = make_reduction(
+                regressor, window_length=int(12), strategy="recursive"
+            )
+            forecaster.fit(y_train)
+            y_pred = forecaster.predict(fh)
+            recursive_without = mean_absolute_percentage_error(
+                y_test, y_pred, symmetric=False
+            )
+            forecaster = make_reduction(
+                regressor,
+                window_length=None,
+                strategy="recursive",
+                transformers=[WindowSummarizer(**kwargs, n_jobs=1)],
+                pooling="global",
+            )
+
+            forecaster.fit(y_train)
+            y_pred = forecaster.predict(fh)
+            recursive_global = mean_absolute_percentage_error(
+                y_test, y_pred, symmetric=False
+            )
+            np.testing.assert_almost_equal(recursive_without, recursive_global)