Fix frontdoor estimation bug

dowhy/causal_estimators/two_stage_regression_estimator.py

@@ -125,7 +125,6 @@ def __init__(
                 )
             )
         else:
-            modified_target_estimand = copy.deepcopy(self._target_estimand)
             self._second_stage_model = self.__class__.DEFAULT_SECOND_STAGE_MODEL(
                 modified_target_estimand,
                 test_significance=self._significance_test,

tests/causal_estimators/test_two_stage_regression_estimator.py

@@ -1,6 +1,9 @@
+import numpy as np
+import pandas as pd
 import pytest
 from pytest import mark
 
+from dowhy import CausalModel
 from dowhy.causal_estimators.two_stage_regression_estimator import TwoStageRegressionEstimator
 
 from .base import TestEstimator
@@ -71,3 +74,72 @@ def test_average_treatment_effect(
             ],
             method_params={"num_simulations": 10, "num_null_simulations": 10},
         )
+
+    def test_frontdoor_estimator(self):
+        """
+        Test for frontdoor estimation, from @AlxndrMlk
+        See issue #616 https://github.com/py-why/dowhy/issues/616
+        """
+
+        # Create the graph describing the causal structure
+        graph = """
+        graph [
+            directed 1
+
+            node [
+                id "X" 
+                label "X"
+            ]    
+            node [
+                id "Z"
+                label "Z"
+            ]
+            node [
+                id "Y"
+                label "Y"
+            ]
+            node [
+                id "U"
+                label "U"
+            ]
+
+            edge [
+                source "X"
+                target "Z"
+            ]
+
+            edge [
+                source "Z"
+                target "Y"
+            ]
+
+            edge [
+                source "U"
+                target "Y"
+            ]
+
+            edge [
+                source "U"
+                target "X"
+            ]
+        ]
+        """.replace(
+            "\n", ""
+        )
+
+        N_SAMPLES = 10000
+        # Generate the data
+        U = np.random.randn(N_SAMPLES)
+        X = np.random.randn(N_SAMPLES) + 0.3 * U
+        Z = 0.7 * X + 0.3 * np.random.randn(N_SAMPLES)
+        Y = 0.65 * Z + 0.2 * U
+
+        # Data to df
+        df = pd.DataFrame(np.vstack([X, Z, Y]).T, columns=["X", "Z", "Y"])
+
+        # Create a model
+        model = CausalModel(data=df, treatment="X", outcome="Y", graph=graph)
+        estimand = model.identify_effect(proceed_when_unidentifiable=True)
+        # Estimate the effect with front-door
+        estimate = model.estimate_effect(identified_estimand=estimand, method_name="frontdoor.two_stage_regression")
+        assert estimate.value == pytest.approx(0.45, 0.025)