APO Models
APO Pointwise Coverage
The simulations are based on the the make_irm_data_discrete_treatments-DGP with \(500\) observations. Due to the linearity of the DGP, Lasso and Logit Regression are nearly optimal choices for the nuisance estimation.
Metadata
DoubleML Version 0.10.dev0
Script irm_apo_coverage.py
Date 2025-01-08 15:02:49
Total Runtime (seconds) 10054.695461
Python Version 3.12.8| Learner g | Learner m | Treatment Level | Bias | CI Length | Coverage |
|---|---|---|---|---|---|
| LGBM | LGBM | 0.000 | 2.076 | 10.316 | 0.963 |
| LGBM | LGBM | 1.000 | 9.193 | 45.558 | 0.967 |
| LGBM | LGBM | 2.000 | 9.594 | 44.674 | 0.952 |
| LGBM | Logistic | 0.000 | 1.337 | 6.704 | 0.955 |
| LGBM | Logistic | 1.000 | 1.695 | 8.845 | 0.968 |
| LGBM | Logistic | 2.000 | 1.661 | 8.724 | 0.971 |
| Linear | LGBM | 0.000 | 1.307 | 6.552 | 0.949 |
| Linear | LGBM | 1.000 | 2.127 | 12.751 | 0.982 |
| Linear | LGBM | 2.000 | 1.635 | 8.953 | 0.967 |
| Linear | Logistic | 0.000 | 1.286 | 6.358 | 0.951 |
| Linear | Logistic | 1.000 | 1.280 | 6.455 | 0.954 |
| Linear | Logistic | 2.000 | 1.281 | 6.394 | 0.959 |
| Learner g | Learner m | Treatment Level | Bias | CI Length | Coverage |
|---|---|---|---|---|---|
| LGBM | LGBM | 0.000 | 2.076 | 8.658 | 0.911 |
| LGBM | LGBM | 1.000 | 9.193 | 38.233 | 0.914 |
| LGBM | LGBM | 2.000 | 9.594 | 37.492 | 0.891 |
| LGBM | Logistic | 0.000 | 1.337 | 5.626 | 0.901 |
| LGBM | Logistic | 1.000 | 1.695 | 7.423 | 0.920 |
| LGBM | Logistic | 2.000 | 1.661 | 7.321 | 0.918 |
| Linear | LGBM | 0.000 | 1.307 | 5.498 | 0.901 |
| Linear | LGBM | 1.000 | 2.127 | 10.701 | 0.951 |
| Linear | LGBM | 2.000 | 1.635 | 7.514 | 0.932 |
| Linear | Logistic | 0.000 | 1.286 | 5.336 | 0.900 |
| Linear | Logistic | 1.000 | 1.280 | 5.418 | 0.904 |
| Linear | Logistic | 2.000 | 1.281 | 5.366 | 0.905 |
APOS Coverage
The simulations are based on the the make_irm_data_discrete_treatments-DGP with \(500\) observations. Due to the linearity of the DGP, Lasso and Logit Regression are nearly optimal choices for the nuisance estimation.
The non-uniform results (coverage, ci length and bias) refer to averaged values over all quantiles (point-wise confidende intervals).
Metadata
DoubleML Version 0.10.dev0
Script irm_apo_coverage.py
Date 2025-01-08 15:02:49
Total Runtime (seconds) 10054.695461
Python Version 3.12.8| Learner g | Learner m | Bias | CI Length | Coverage | Uniform CI Length | Uniform Coverage |
|---|---|---|---|---|---|---|
| LGBM | LGBM | 7.118 | 33.615 | 0.958 | 40.868 | 0.973 |
| LGBM | Logistic | 1.571 | 8.090 | 0.963 | 9.592 | 0.959 |
| Linear | LGBM | 1.726 | 9.418 | 0.967 | 11.234 | 0.974 |
| Linear | Logistic | 1.280 | 6.402 | 0.957 | 6.818 | 0.953 |
| Learner g | Learner m | Bias | CI Length | Coverage | Uniform CI Length | Uniform Coverage |
|---|---|---|---|---|---|---|
| LGBM | LGBM | 7.118 | 28.210 | 0.906 | 36.194 | 0.925 |
| LGBM | Logistic | 1.571 | 6.789 | 0.915 | 8.394 | 0.924 |
| Linear | LGBM | 1.726 | 7.903 | 0.927 | 9.850 | 0.939 |
| Linear | Logistic | 1.280 | 5.373 | 0.904 | 5.796 | 0.906 |
Causal Contrast Coverage
The simulations are based on the the make_irm_data_discrete_treatments-DGP with \(500\) observations. Due to the linearity of the DGP, Lasso and Logit Regression are nearly optimal choices for the nuisance estimation.
The non-uniform results (coverage, ci length and bias) refer to averaged values over all quantiles (point-wise confidende intervals).
Metadata
DoubleML Version 0.10.dev0
Script irm_apo_coverage.py
Date 2025-01-08 15:02:49
Total Runtime (seconds) 10054.695461
Python Version 3.12.8| Learner g | Learner m | Bias | CI Length | Coverage | Uniform CI Length | Uniform Coverage |
|---|---|---|---|---|---|---|
| LGBM | LGBM | 9.786 | 45.131 | 0.949 | 51.424 | 0.965 |
| LGBM | Logistic | 1.268 | 6.822 | 0.962 | 7.765 | 0.961 |
| Linear | LGBM | 1.507 | 8.855 | 0.989 | 10.090 | 0.992 |
| Linear | Logistic | 0.297 | 1.361 | 0.933 | 1.550 | 0.916 |
| Learner g | Learner m | Bias | CI Length | Coverage | Uniform CI Length | Uniform Coverage |
|---|---|---|---|---|---|---|
| LGBM | LGBM | 9.786 | 37.875 | 0.888 | 44.829 | 0.898 |
| LGBM | Logistic | 1.268 | 5.725 | 0.927 | 6.774 | 0.926 |
| Linear | LGBM | 1.507 | 7.431 | 0.958 | 8.799 | 0.975 |
| Linear | Logistic | 0.297 | 1.143 | 0.873 | 1.351 | 0.869 |