Coverage simulations for LPLR

.gitignore

@@ -1,4 +1,5 @@
 __pycache__/
+.idea/
 
 # Logs
 monte-cover/logs/

doc/_quarto-dev.yml

@@ -21,6 +21,7 @@ website:
     - plm/plr_gate.qmd
     - plm/plr_cate.qmd
     - plm/pliv.qmd
+    - plm/lplr.qmd
     # DID
     - did/did_pa.qmd
     - did/did_cs.qmd

doc/_website.yml

@@ -25,6 +25,7 @@ website:
           - plm/plr_gate.qmd
           - plm/plr_cate.qmd
           - plm/pliv.qmd
+          - plm/lplr.qmd
       - text: "DID"
         menu:
           - did/did_pa_multi.qmd

doc/plm/lplr.qmd

@@ -0,0 +1,113 @@
+---
+title: "Logistic Partial Linear Regression Models"
+
+jupyter: python3
+---
+
+```{python}
+#| echo: false
+
+import numpy as np
+import pandas as pd
+from itables import init_notebook_mode
+import os
+import sys
+
+doc_dir = os.path.abspath(os.path.join(os.getcwd(), ".."))
+if doc_dir not in sys.path:
+    sys.path.append(doc_dir)
+
+from utils.style_tables import generate_and_show_styled_table
+
+init_notebook_mode(all_interactive=True)
+```
+
+## ATE Coverage
+
+The simulations are based on the  the [make_lplr_LZZ2020](https://docs.doubleml.org/stable/api/generated/doubleml.datasets.make_lplr_LZZ2020.html)-DGP with $500$ observations.
+
+::: {.callout-note title="Metadata"  collapse="true"}
+
+```{python}
+#| echo: false
+metadata_file = '../../results/plm/lplr_ate_metadata.csv'
+metadata_df = pd.read_csv(metadata_file)
+print(metadata_df.T.to_string(header=False))
+```
+
+:::
+
+```{python}
+#| echo: false
+
+# set up data and rename columns
+df_coverage = pd.read_csv("../../results/plm/lplr_ate_coverage.csv", index_col=None)
+
+if "repetition" in df_coverage.columns and df_coverage["repetition"].nunique() == 1:
+    n_rep_coverage = df_coverage["repetition"].unique()[0]
+elif "n_rep" in df_coverage.columns and df_coverage["n_rep"].nunique() == 1:
+    n_rep_coverage = df_coverage["n_rep"].unique()[0]
+else:
+    n_rep_coverage = "N/A" # Fallback if n_rep cannot be determined
+
+display_columns_coverage = ["Learner m", "Learner M", "Learner t", "Bias", "CI Length", "Coverage"]
+```
+
+### Nuisance space
+
+```{python}
+# | echo: false
+
+generate_and_show_styled_table(
+    main_df=df_coverage,
+    filters={"level": 0.95, "Score": "nuisance_space"},
+    display_cols=display_columns_coverage,
+    n_rep=n_rep_coverage,
+    level_col="level",
+#    rename_map={"Learner g": "Learner l"},
+    coverage_highlight_cols=["Coverage"]
+)
+```
+
+```{python}
+#| echo: false
+
+generate_and_show_styled_table(
+    main_df=df_coverage,
+    filters={"level": 0.9, "Score": "nuisance_space"},
+    display_cols=display_columns_coverage,
+    n_rep=n_rep_coverage,
+    level_col="level",
+#    rename_map={"Learner g": "Learner l"},
+    coverage_highlight_cols=["Coverage"]
+)
+```
+
+### Instrument
+
+
+```{python}
+#| echo: false
+
+generate_and_show_styled_table(
+    main_df=df_coverage,
+    filters={"level": 0.95, "Score": "instrument"},
+    display_cols=display_columns_coverage,
+    n_rep=n_rep_coverage,
+    level_col="level",
+    coverage_highlight_cols=["Coverage"]
+)
+```
+
+```{python}
+#| echo: false
+
+generate_and_show_styled_table(
+    main_df=df_coverage,
+    filters={"level": 0.9, "Score": "instrument"},
+    display_cols=display_columns_coverage,
+    n_rep=n_rep_coverage,
+    level_col="level",
+    coverage_highlight_cols=["Coverage"]
+)
+```

monte-cover/src/montecover/plm/__init__.py

@@ -5,11 +5,13 @@
 from montecover.plm.plr_ate_sensitivity import PLRATESensitivityCoverageSimulation
 from montecover.plm.plr_cate import PLRCATECoverageSimulation
 from montecover.plm.plr_gate import PLRGATECoverageSimulation
+from montecover.plm.lplr_ate import LPLRATECoverageSimulation
 
 __all__ = [
     "PLRATECoverageSimulation",
     "PLIVLATECoverageSimulation",
     "PLRGATECoverageSimulation",
     "PLRCATECoverageSimulation",
     "PLRATESensitivityCoverageSimulation",
+    "LPLRATECoverageSimulation",
 ]

monte-cover/src/montecover/plm/lplr_ate.py

@@ -0,0 +1,126 @@
+import warnings
+from typing import Any, Dict, Optional
+
+import doubleml as dml
+from doubleml.plm.datasets import make_lplr_LZZ2020
+
+from montecover.base import BaseSimulation
+from montecover.utils import create_learner_from_config
+
+
+class LPLRATECoverageSimulation(BaseSimulation):
+    """Simulation class for coverage properties of DoubleMLPLR for ATE estimation."""
+
+    def __init__(
+            self,
+            config_file: str,
+            suppress_warnings: bool = True,
+            log_level: str = "INFO",
+            log_file: Optional[str] = None,
+            use_failed_scores: bool = False,
+    ):
+        super().__init__(
+            config_file=config_file,
+            suppress_warnings=suppress_warnings,
+            log_level=log_level,
+            log_file=log_file,
+        )
+
+        # Calculate oracle values
+        self._calculate_oracle_values()
+
+        self._use_failed_scores = use_failed_scores
+
+    def _process_config_parameters(self):
+        """Process simulation-specific parameters from config"""
+        # Process ML models in parameter grid
+        assert "learners" in self.dml_parameters, "No learners specified in the config file"
+
+        required_learners = ["ml_m", "ml_M", "ml_t"]
+        for learner in self.dml_parameters["learners"]:
+            for ml in required_learners:
+                assert ml in learner, f"No {ml} specified in the config file"
+
+    def _calculate_oracle_values(self):
+        """Calculate oracle values for the simulation."""
+        self.logger.info("Calculating oracle values")
+
+        self.oracle_values = dict()
+        self.oracle_values["theta"] = self.dgp_parameters["theta"]
+
+    def run_single_rep(self, dml_data, dml_params) -> Dict[str, Any]:
+        """Run a single repetition with the given parameters."""
+        # Extract parameters
+        learner_config = dml_params["learners"]
+        learner_m_name, ml_m = create_learner_from_config(learner_config["ml_m"])
+        learner_M_name, ml_M = create_learner_from_config(learner_config["ml_M"])
+        learner_t_name, ml_t = create_learner_from_config(learner_config["ml_t"])
+        score = dml_params["score"]
+
+        # Model
+        dml_model = dml.DoubleMLLPLR(
+            obj_dml_data=dml_data,
+            ml_m=ml_m,
+            ml_M=ml_M,
+            ml_t=ml_t,
+            score=score,
+            error_on_convergence_failure= not self._use_failed_scores,)
+
+        try:
+            dml_model.fit()
+        except RuntimeError as e:
+            self.logger.info(f"Exception during fit: {e}")
+            return None
+
+        result = {
+            "coverage": [],
+        }
+        for level in self.confidence_parameters["level"]:
+            level_result = dict()
+            level_result["coverage"] = self._compute_coverage(
+                thetas=dml_model.coef,
+                oracle_thetas=self.oracle_values["theta"],
+                confint=dml_model.confint(level=level),
+                joint_confint=None,
+            )
+
+            # add parameters to the result
+            for res in level_result.values():
+                res.update(
+                    {
+                        "Learner m": learner_m_name,
+                        "Learner M": learner_M_name,
+                        "Learner t": learner_t_name,
+                        "Score": score,
+                        "level": level,
+                    }
+                )
+            for key, res in level_result.items():
+                result[key].append(res)
+
+        return result
+
+    def summarize_results(self):
+        """Summarize the simulation results."""
+        self.logger.info("Summarizing simulation results")
+
+        # Group by parameter combinations
+        groupby_cols = ["Learner m", "Learner M", "Learner t", "Score", "level"]
+        aggregation_dict = {
+            "Coverage": "mean",
+            "CI Length": "mean",
+            "Bias": "mean",
+            "repetition": "count",
+        }
+
+        # Aggregate results (possibly multiple result dfs)
+        result_summary = dict()
+        for result_name, result_df in self.results.items():
+            result_summary[result_name] = result_df.groupby(groupby_cols).agg(aggregation_dict).reset_index()
+            self.logger.debug(f"Summarized {result_name} results")
+
+        return result_summary
+
+    def _generate_dml_data(self, dgp_params) -> dml.DoubleMLData:
+        """Generate data for the simulation."""
+        return make_lplr_LZZ2020(**dgp_params)

results/plm/logistic_ate_config.yml

@@ -0,0 +1,91 @@
+simulation_parameters:
+  repetitions: 1000
+  max_runtime: 86400
+  random_seed: 42
+  n_jobs: -2
+dgp_parameters:
+  theta:
+  - 0.5
+  n_obs:
+  - 500
+  dim_x:
+  - 20
+learner_definitions:
+  lasso: &id001
+    name: LassoCV
+  logistic: &id002
+    name: Logistic
+  rf: &id003
+    name: RF Regr.
+    params:
+      n_estimators: 100
+      max_features: sqrt
+  rf-class: &id004
+    name: RF Clas.
+    params:
+      n_estimators: 100
+      max_features: sqrt
+  lgbm: &id005
+    name: LGBM Regr.
+    params:
+      n_estimators: 500
+      learning_rate: 0.01
+  lgbm-class: &id006
+    name: LGBM Clas.
+    params:
+      n_estimators: 500
+      learning_rate: 0.01
+dml_parameters:
+  learners:
+  - ml_m: *id001
+    ml_M: *id002
+    ml_t: *id001
+  - ml_m: *id003
+    ml_M: *id004
+    ml_t: *id003
+  - ml_m: *id005
+    ml_M: *id006
+    ml_t: *id005
+  - ml_m: *id003
+    ml_M: *id006
+    ml_t: *id005
+  - ml_m: *id005
+    ml_M: *id004
+    ml_t: *id005
+  - ml_m: *id005
+    ml_M: *id006
+    ml_t: *id003
+  - ml_m: *id005
+    ml_M: *id004
+    ml_t: *id003
+  - ml_m: *id003
+    ml_M: *id006
+    ml_t: *id003
+  - ml_m: *id003
+    ml_M: *id004
+    ml_t: *id005
+  - ml_m: *id001
+    ml_M: *id006
+    ml_t: *id005
+  - ml_m: *id005
+    ml_M: *id002
+    ml_t: *id005
+  - ml_m: *id005
+    ml_M: *id006
+    ml_t: *id001
+  - ml_m: *id001
+    ml_M: *id004
+    ml_t: *id003
+  - ml_m: *id003
+    ml_M: *id002
+    ml_t: *id003
+  - ml_m: *id003
+    ml_M: *id004
+    ml_t: *id001
+  score:
+  - nuisance_space
+  - instrument
+confidence_parameters:
+  level:
+  - 0.95
+  - 0.9