Add RDD Coverage Simulation

.github/workflows/apo_sim.yml

@@ -59,7 +59,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

.github/workflows/did_sim.yml

@@ -60,7 +60,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

.github/workflows/iivm_sim.yml

@@ -59,7 +59,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

.github/workflows/irm_sim.yml

@@ -64,7 +64,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

.github/workflows/pliv_sim.yml

@@ -59,7 +59,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

.github/workflows/plr_sim.yml

@@ -62,6 +62,7 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
+        pip uninstall -y doubleml
         pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
 
     - name: Set up Git configuration

.github/workflows/quant_sim.yml

@@ -61,7 +61,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

.github/workflows/rdd_sim.yml

@@ -12,7 +12,7 @@ on:
     types: [run-rdd-scripts]
 
 jobs:
-  run-apo-scripts:
+  run-rdd-scripts:
     runs-on: ubuntu-latest
     strategy:
       matrix:
@@ -60,7 +60,14 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml[rdd] @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
+
+    - name: Install RDFlex from main branch
+      run: |
+        pip uninstall -y doubleml
+        pip install git+https://github.com/DoubleML/doubleml-rdflex.git@main
+        pip install rdrobust
 
     - name: Set up Git configuration
       run: |
@@ -72,7 +79,7 @@ jobs:
 
     - name: Commit any existing changes
       run: |
-        git add results/irm
+        git add results/rdd
         git commit -m "Update results from script: ${{ matrix.script }}" || echo "No changed results to commit"
 
     - name: Wait random time

.github/workflows/sim.yml

@@ -45,3 +45,14 @@ jobs:
 
           echo "HTTP Status: $RESPONSE"
           cat response.txt
+
+      - name: Trigger RDD Scripts Workflows
+        run: |
+          RESPONSE=$(curl -s -o response.txt -w "%{http_code}" -X POST \
+            -H "Authorization: token ${{ secrets.GITHUB_TOKEN }}" \
+            -H "Accept: application/vnd.github.everest-preview+json" \
+            https://api.github.com/repos/${{ github.repository }}/dispatches \
+            -d '{"event_type": "run-rdd-scripts", "client_payload": { "doubleml_py_branch": "${{ github.event.inputs.doubleml-py-branch }}", "target_branch": "${{ github.ref_name }}"}}')
+
+          echo "HTTP Status: $RESPONSE"
+          cat response.txt

.github/workflows/ssm_sim.yml

@@ -60,7 +60,8 @@ jobs:
 
     - name: Install DoubleML from correct branch
       run: |
-        pip install git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}
+        pip uninstall -y doubleml
+        pip install "doubleml @ git+https://github.com/DoubleML/doubleml-for-py@${{ env.DML_BRANCH }}"
 
     - name: Set up Git configuration
       run: |

doc/_website.yml

@@ -1,7 +1,7 @@
 website:
   title: "DoubleML Coverage"
   favicon: _static/favicon.ico
-  search: 
+  search:
     location: sidebar
   sidebar:
     style: "docked"
@@ -18,20 +18,23 @@ website:
           - irm/apo.qmd
           - irm/qte.qmd
       - text: "PLM"
-        menu:      
+        menu:
           - plm/plr.qmd
           - plm/plr_gate.qmd
           - plm/plr_cate.qmd
           - plm/pliv.qmd
       - text: "DID"
-        menu:      
+        menu:
           - did/did_pa.qmd
           - did/did_cs.qmd
       - text: "SSM"
-        menu:      
+        menu:
           - ssm/ssm_mar.qmd
           - ssm/ssm_nonignorable.qmd
-    right: 
+      - text: "RDD"
+        menu:
+          - rdd/rdd.qmd
+    right:
       - icon: book
         href: https://docs.doubleml.org/stable/index.html
         aria-label: Documentation
@@ -43,9 +46,8 @@ website:
             href: https://github.com/DoubleML/doubleml-coverage
         aria-label: GitHub
 
-
 format:
   html:
-    theme: cosmo 
+    theme: cosmo
     css: styles.css
     toc: true

doc/rdd/rdd.qmd

@@ -0,0 +1,149 @@
+---
+title: "Flexible covariate adjustments in RDD"
+
+jupyter: python3
+---
+
+
+```{python}
+#| echo: false
+
+import numpy as np
+import pandas as pd
+from itables import init_notebook_mode, show, options
+
+init_notebook_mode(all_interactive=True)
+
+def highlight_range(s, level=0.95, dist=0.05, props=''):
+    color_grid = np.where((s >= level-dist) &
+                          (s <= level+dist), props, '')
+    return color_grid
+
+
+def color_coverage(df, level):
+    # color coverage column order is important
+    styled_df = df.apply(
+        highlight_range,
+        level=level,
+        dist=1.0,
+        props='color:black;background-color:red',
+        subset=["Coverage"])
+    styled_df = styled_df.apply(
+        highlight_range,
+        level=level,
+        dist=0.1,
+        props='color:black;background-color:yellow',
+        subset=["Coverage"])
+    styled_df = styled_df.apply(
+        highlight_range,
+        level=level,
+        dist=0.05,
+        props='color:white;background-color:darkgreen',
+        subset=["Coverage"])
+
+    # set all coverage values to bold
+    styled_df = styled_df.set_properties(
+        **{'font-weight': 'bold'},
+        subset=["Coverage"])
+    return styled_df
+
+
+def make_pretty(df, level, n_rep):
+    styled_df = df.style.hide(axis="index")
+    # Format only float columns
+    float_cols = df.select_dtypes(include=['float']).columns
+    styled_df = styled_df.format({col: "{:.3f}" for col in float_cols})
+
+    # color coverage column order is important
+    styled_df = color_coverage(styled_df, level)
+    caption = f"Coverage for {level*100}%-Confidence Interval over {n_rep} Repetitions"
+
+    return show(styled_df, caption=caption)
+```
+
+## Sharp Design
+
+The simulations are based on the [make_simple_rdd_data](https://docs.doubleml.org/stable/api/generated/doubleml.rdd.datasets.make_simple_rdd_data.html#doubleml.rdd.datasets.make_simple_rdd_data)-DGP with $1000$ observations. The simulation considers data under a [sharp regression discontinuity design](https://docs.doubleml.org/stable/guide/models.html#sharp-regression-discontinuity-design).
+
+::: {.callout-note title="Metadata"  collapse="true"}
+
+```{python}
+#| echo: false
+metadata_file = '../../results/rdd/rdd_sharp_coverage_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 = pd.read_csv("../../results/rdd/rdd_sharp_coverage.csv", index_col=None)
+
+assert df["repetition"].nunique() == 1
+n_rep = df["repetition"].unique()[0]
+
+display_columns = ["Method", "Learner g", "fs specification", "Bias", "CI Length", "Coverage"]
+```
+
+```{python}
+#| echo: false
+level = 0.95
+
+df_ate_95 = df[(df['level'] == level)][display_columns]
+make_pretty(df_ate_95, level, n_rep)
+```
+
+```{python}
+#| echo: false
+level = 0.9
+
+df_ate_9 = df[(df['level'] == level)][display_columns]
+make_pretty(df_ate_9, level, n_rep)
+```
+
+
+## Fuzzy Design
+
+The simulations are based on the [make_simple_rdd_data](https://docs.doubleml.org/stable/api/generated/doubleml.rdd.datasets.make_simple_rdd_data.html#doubleml.rdd.datasets.make_simple_rdd_data)-DGP with $2000$ observations. The simulation considers data under a [fuzzy regression discontinuity design](https://docs.doubleml.org/stable/guide/models.html#fuzzy-regression-discontinuity-design).
+
+::: {.callout-note title="Metadata"  collapse="true"}
+
+```{python}
+#| echo: false
+metadata_file = '../../results/rdd/rdd_fuzzy_coverage_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 = pd.read_csv("../../results/rdd/rdd_fuzzy_coverage.csv", index_col=None)
+
+assert df["repetition"].nunique() == 1
+n_rep = df["repetition"].unique()[0]
+
+display_columns = ["Method", "Learner g", "Learner m", "fs specification", "Bias", "CI Length", "Coverage"]
+```
+
+```{python}
+#| echo: false
+level = 0.95
+
+df_ate_95 = df[(df['level'] == level)][display_columns]
+make_pretty(df_ate_95, level, n_rep)
+```
+
+```{python}
+#| echo: false
+level = 0.9
+
+df_ate_9 = df[(df['level'] == level)][display_columns]
+make_pretty(df_ate_9, level, n_rep)
+```

doc/ssm/ssm_mar.qmd

@@ -62,7 +62,7 @@ def make_pretty(df, level, n_rep):
 
 ## ATE Coverage
 
-The simulations are based on the  the [make_ssm_data](https://docs.doubleml.org/stable/api/generated/doubleml.datasets.make_ssm_data.html)-DGP with $500$ observations. The simulation considers data under [missingness at random](https://docs.doubleml.org/stable/guide/models.html#missingness-at-random).
+The simulations are based on the [make_ssm_data](https://docs.doubleml.org/stable/api/generated/doubleml.datasets.make_ssm_data.html)-DGP with $500$ observations. The simulation considers data under [missingness at random](https://docs.doubleml.org/stable/guide/models.html#missingness-at-random).
 
 ::: {.callout-note title="Metadata"  collapse="true"}
 

doc/ssm/ssm_nonignorable.qmd

@@ -62,7 +62,7 @@ def make_pretty(df, level, n_rep):
 
 ## ATE Coverage
 
-The simulations are based on the  the [make_ssm_data](https://docs.doubleml.org/stable/api/generated/doubleml.datasets.make_ssm_data.html)-DGP with $500$ observations. The simulation considers data with [nonignorable nonresponse](https://docs.doubleml.org/stable/guide/models.html#nonignorable-nonresponse).
+The simulations are based on the [make_ssm_data](https://docs.doubleml.org/stable/api/generated/doubleml.datasets.make_ssm_data.html)-DGP with $500$ observations. The simulation considers data with [nonignorable nonresponse](https://docs.doubleml.org/stable/guide/models.html#nonignorable-nonresponse).
 
 ::: {.callout-note title="Metadata"  collapse="true"}
 

requirements.txt

@@ -1,4 +1,4 @@
-doubleml
+doubleml[rdd]
 joblib
 numpy
 pandas

results/did/did_cs_atte_coverage_metadata.csv

@@ -1,2 +1,2 @@
 DoubleML Version,Script,Date,Total Runtime (seconds),Python Version
-0.9.0,did_cs_atte_coverage.py,2024-09-09 12:16:59,17588.891700983047,3.12.5
+0.10.dev0,did_cs_atte_coverage.py,2025-01-08 16:52:06,16598.133431196213,3.12.8

results/did/did_pa_atte_coverage_metadata.csv

@@ -1,2 +1,2 @@
 DoubleML Version,Script,Date,Total Runtime (seconds),Python Version
-0.9.0,did_pa_atte_coverage.py,2024-09-09 11:44:15,15625.949723243713,3.12.5
+0.10.dev0,did_pa_atte_coverage.py,2025-01-08 16:22:26,14811.1988697052,3.12.8

results/irm/cvar_coverage_metadata.csv

@@ -1,2 +1,2 @@
 DoubleML Version,Script,Date,Total Runtime (seconds),Python Version
-0.9.0,cvar_coverage.py,2024-09-09 11:59:26,16546.96124267578,3.12.5
+0.10.dev0,cvar_coverage.py,2025-01-08 16:49:01,16412.1168384552,3.12.8

results/irm/cvar_coverage_pq0.csv

@@ -1,9 +1,9 @@
 Learner g,Learner m,level,Coverage,CI Length,Bias,repetition
-LGBM,LGBM,0.9,0.8876923076923078,0.5699689541680535,0.13934148600639645,100
-LGBM,LGBM,0.95,0.9461538461538461,0.6791598985897508,0.13934148600639645,100
-LGBM,Logistic Regression,0.9,0.8192307692307692,0.4060118886357097,0.1185635852924032,100
-LGBM,Logistic Regression,0.95,0.89,0.48379300503226824,0.1185635852924032,100
-Linear,LGBM,0.9,0.77,0.5801661718639771,0.17503081399943532,100
-Linear,LGBM,0.95,0.8607692307692308,0.691310632915921,0.17503081399943532,100
-Linear,Logistic Regression,0.9,0.6892307692307692,0.4294636844188598,0.15444826651283644,100
-Linear,Logistic Regression,0.95,0.7776923076923077,0.5117375432906374,0.15444826651283644,100
+LGBM,LGBM,0.9,0.8884615384615384,0.5699689541680535,0.13934954721406115,100
+LGBM,LGBM,0.95,0.9461538461538461,0.6791598985897508,0.13934954721406115,100
+LGBM,Logistic Regression,0.9,0.8192307692307692,0.4060217020758364,0.11848308187809455,100
+LGBM,Logistic Regression,0.95,0.8892307692307692,0.48380469846741503,0.11848308187809455,100
+Linear,LGBM,0.9,0.7692307692307692,0.5801661718639771,0.1751137266106609,100
+Linear,LGBM,0.95,0.8615384615384616,0.691310632915921,0.1751137266106609,100
+Linear,Logistic Regression,0.9,0.6892307692307692,0.42946981193983413,0.1545041280672664,100
+Linear,Logistic Regression,0.95,0.7776923076923077,0.5117448446822186,0.1545041280672664,100

results/irm/cvar_coverage_pq1.csv

@@ -1,9 +1,9 @@
 Learner g,Learner m,level,Coverage,CI Length,Bias,repetition
-LGBM,LGBM,0.9,0.9323076923076923,0.1908311409902022,0.04347081995558675,100
-LGBM,LGBM,0.95,0.9776923076923077,0.22738932956769187,0.04347081995558675,100
-LGBM,Logistic Regression,0.9,0.9146153846153847,0.1776482729176802,0.04405662937860094,100
-LGBM,Logistic Regression,0.95,0.9646153846153847,0.21168097338831957,0.04405662937860094,100
-Linear,LGBM,0.9,0.9307692307692308,0.21584587388672186,0.04750929738761366,100
-Linear,LGBM,0.95,0.9761538461538461,0.25719622226423833,0.04750929738761366,100
-Linear,Logistic Regression,0.9,0.8853846153846153,0.1934225485324801,0.04921876802543021,100
-Linear,Logistic Regression,0.95,0.943076923076923,0.23047718210904136,0.04921876802543021,100
+LGBM,LGBM,0.9,0.9323076923076923,0.1908311409902022,0.04358145932430658,100
+LGBM,LGBM,0.95,0.9776923076923077,0.22738932956769187,0.04358145932430658,100
+LGBM,Logistic Regression,0.9,0.9130769230769231,0.17764550716816235,0.04414608757632295,100
+LGBM,Logistic Regression,0.95,0.963076923076923,0.21167767779450106,0.04414608757632295,100
+Linear,LGBM,0.9,0.9292307692307692,0.21584587388672186,0.047457925834839765,100
+Linear,LGBM,0.95,0.9769230769230769,0.25719622226423833,0.047457925834839765,100
+Linear,Logistic Regression,0.9,0.8823076923076922,0.19342205011642188,0.0494625033437771,100
+Linear,Logistic Regression,0.95,0.9415384615384617,0.2304765882096771,0.0494625033437771,100