Fixed examples

Author: jmafoster1

examples/.gitignore

@@ -0,0 +1 @@
+*.pdf

examples/covasim_/doubling_beta/causal_test_beta.py

@@ -47,19 +47,19 @@ def doubling_beta_CATE_on_csv(observational_data_path: str, simulate_counterfact
     past_execution_df = pd.read_csv(observational_data_path)
     _, causal_test_engine, causal_test_case = engine_setup(observational_data_path)
 
-    linear_regression_estimator = LinearRegressionEstimator(('beta',), 0.032, 0.016,
+    linear_regression_estimator = LinearRegressionEstimator('beta', 0.032, 0.016,
                                                             {'avg_age', 'contacts'},  # We use custom adjustment set
-                                                            ('cum_infections',),
+                                                            'cum_infections',
                                                             df=past_execution_df)
 
     # Add squared terms for beta, since it has a quadratic relationship with cumulative infections
     linear_regression_estimator.add_squared_term_to_df('beta')
     causal_test_result = causal_test_engine.execute_test(linear_regression_estimator, causal_test_case, 'ate')
 
     # Repeat for association estimate (no adjustment)
-    no_adjustment_linear_regression_estimator = LinearRegressionEstimator(('beta',), 0.032, 0.016,
+    no_adjustment_linear_regression_estimator = LinearRegressionEstimator('beta', 0.032, 0.016,
                                                                           set(),
-                                                                          ('cum_infections',),
+                                                                          'cum_infections',
                                                                           df=past_execution_df)
     no_adjustment_linear_regression_estimator.add_squared_term_to_df('beta')
     association_test_result = causal_test_engine.execute_test(no_adjustment_linear_regression_estimator, causal_test_case, 'ate')
@@ -79,9 +79,9 @@ def doubling_beta_CATE_on_csv(observational_data_path: str, simulate_counterfact
     # Repeat causal inference after deleting all rows with treatment value to obtain counterfactual inferences
     if simulate_counterfactuals:
         counterfactual_past_execution_df = past_execution_df[past_execution_df['beta'] != 0.032]
-        counterfactual_linear_regression_estimator = LinearRegressionEstimator(('beta',), 0.032, 0.016,
+        counterfactual_linear_regression_estimator = LinearRegressionEstimator('beta', 0.032, 0.016,
                                                                                {'avg_age', 'contacts'},
-                                                                               ('cum_infections',),
+                                                                               'cum_infections',
                                                                                df=counterfactual_past_execution_df)
         counterfactual_linear_regression_estimator.add_squared_term_to_df('beta')
         counterfactual_causal_test_result = causal_test_engine.execute_test(linear_regression_estimator, causal_test_case, 'ate')

examples/covasim_/vaccinating_elderly/causal_test_vaccine.py

@@ -85,9 +85,9 @@ def experimental_causal_test_vaccinate_elderly(runs_per_test_per_config: int = 3
         minimal_adjustment_set = causal_dag.identification(base_test_case)
 
         # 9. Build statistical model
-        linear_regression_estimator = LinearRegressionEstimator((vaccine.name,), 1, 0,
+        linear_regression_estimator = LinearRegressionEstimator(vaccine.name, 1, 0,
                                                                 minimal_adjustment_set,
-                                                                (outcome_variable.name,))
+                                                                outcome_variable.name)
 
         # 10. Execute test and save results in dict
         causal_test_result = causal_test_engine.execute_test(linear_regression_estimator, causal_test_case, 'ate')

examples/lr91/causal_test_max_conductances.py

@@ -125,9 +125,9 @@ def effects_on_APD90(observational_data_path, treatment_var, control_val, treatm
 
     # 9. Obtain the minimal adjustment set from the causal DAG
     minimal_adjustment_set = causal_dag.identification(base_test_case)
-    linear_regression_estimator = LinearRegressionEstimator((treatment_var.name,), treatment_val, control_val,
+    linear_regression_estimator = LinearRegressionEstimator(treatment_var.name, treatment_val, control_val,
                                                             minimal_adjustment_set,
-                                                            ('APD90',)
+                                                            'APD90'
                                                             )
 
     # 10. Run the causal test and print results

examples/lr91/causal_test_max_conductances_test_suite.py

@@ -82,7 +82,7 @@ def causal_testing_sensitivity_analysis():
     for base_test_case in causal_test_results:
         # Place results of test_suite into format required for plotting
         results[base_test_case.treatment_variable.name] = \
-            {"ate": [result.ate for result in causal_test_results[base_test_case]['LinearRegressionEstimator']],
+            {"ate": [result.test_value.value for result in causal_test_results[base_test_case]['LinearRegressionEstimator']],
              "cis": [result.confidence_intervals for result in
                      causal_test_results[base_test_case]['LinearRegressionEstimator']]}
 

examples/poisson-line-process/causal_test_poisson.py

@@ -118,11 +118,11 @@ def test_intensity_num_shapes(
         )
     else:
         estimator = LinearRegressionEstimator(
-            treatment=[treatment],
+            treatment=treatment,
             control_value=causal_test_case.control_value,
             treatment_value=causal_test_case.treatment_value,
             adjustment_set=set(),
-            outcome=[outcome],
+            outcome=outcome,
             df=data,
             intercept=0,
             effect_modifiers=causal_test_case.effect_modifier_configuration,

examples/poisson/README.md

@@ -0,0 +1,11 @@
+# Poisson Line Process Case Study: Statistical Metamorphic Testing
+Here we demonstrate how the same test suite as in `poisson-line-process` can be coded using the JSON front end.
+
+## How to run
+To run this case study:
+1. Ensure all project dependencies are installed by running `pip install .` in the top level directory
+   (instructions are provided in the project README).
+2. Change directory to `causal_testing/examples/poisson`.
+3. Run the command `python run_causal_tests.py --data_path data.csv --dag_path dag.dot --json_path causal_tests.json`
+
+This should print a series of causal test results and produce two CSV files. `intensity_num_shapes_results_random_1000.csv` corresponds to table 1, and `width_num_shapes_results_random_1000.csv` relates to our findings regarding the relationship of width and `P_u`.