This is the init_notebook_mode
cell from ITables v2.1.4
(you should not see this message - is your notebook trusted?)
LATE Coverage
The simulations are based on the the make_pliv_CHS2015-DGP with \(500\) observations. Due to the linearity of the DGP, Lasso is a nearly optimal choice for the nuisance estimation.
DoubleML Version 0.8.2
Script pliv_late_coverage.py
Date 2024-08-12 19:43:22
Total Runtime (seconds) 6774.6652
Python Version 3.12.4
Partialling out
Coverage for 95.0%-Confidence Interval over 1000 Repetitions
Lasso |
Lasso |
Lasso |
0.072 |
0.357 |
0.947 |
Lasso |
Lasso |
Random Forest |
0.074 |
0.366 |
0.948 |
Lasso |
Random Forest |
Lasso |
0.077 |
0.378 |
0.956 |
Lasso |
Random Forest |
Random Forest |
0.082 |
0.359 |
0.950 |
Random Forest |
Lasso |
Lasso |
0.081 |
0.394 |
0.953 |
Random Forest |
Lasso |
Random Forest |
0.077 |
0.380 |
0.952 |
Random Forest |
Random Forest |
Lasso |
0.107 |
0.419 |
0.861 |
Random Forest |
Random Forest |
Random Forest |
0.079 |
0.362 |
0.923 |
Coverage for 90.0%-Confidence Interval over 1000 Repetitions
Lasso |
Lasso |
Lasso |
0.072 |
0.299 |
0.899 |
Lasso |
Lasso |
Random Forest |
0.074 |
0.307 |
0.897 |
Lasso |
Random Forest |
Lasso |
0.077 |
0.317 |
0.905 |
Lasso |
Random Forest |
Random Forest |
0.082 |
0.301 |
0.871 |
Random Forest |
Lasso |
Lasso |
0.081 |
0.331 |
0.906 |
Random Forest |
Lasso |
Random Forest |
0.077 |
0.319 |
0.906 |
Random Forest |
Random Forest |
Lasso |
0.107 |
0.351 |
0.786 |
Random Forest |
Random Forest |
Random Forest |
0.079 |
0.304 |
0.868 |
IV-type
For the IV-type score, the learners ml_l
and ml_g
are both set to the same type of learner (here Learner g).
Coverage for 95.0%-Confidence Interval over 1000 Repetitions
Lasso |
Lasso |
Lasso |
0.072 |
0.276 |
0.873 |
Lasso |
Lasso |
Random Forest |
0.072 |
0.276 |
0.866 |
Lasso |
Random Forest |
Lasso |
0.078 |
0.315 |
0.901 |
Lasso |
Random Forest |
Random Forest |
0.076 |
0.316 |
0.915 |
Random Forest |
Lasso |
Lasso |
0.077 |
0.289 |
0.865 |
Random Forest |
Lasso |
Random Forest |
0.075 |
0.288 |
0.875 |
Random Forest |
Random Forest |
Lasso |
0.080 |
0.331 |
0.890 |
Random Forest |
Random Forest |
Random Forest |
0.080 |
0.327 |
0.889 |
Coverage for 90.0%-Confidence Interval over 1000 Repetitions
Lasso |
Lasso |
Lasso |
0.072 |
0.232 |
0.806 |
Lasso |
Lasso |
Random Forest |
0.072 |
0.232 |
0.811 |
Lasso |
Random Forest |
Lasso |
0.078 |
0.265 |
0.828 |
Lasso |
Random Forest |
Random Forest |
0.076 |
0.265 |
0.839 |
Random Forest |
Lasso |
Lasso |
0.077 |
0.242 |
0.794 |
Random Forest |
Lasso |
Random Forest |
0.075 |
0.242 |
0.802 |
Random Forest |
Random Forest |
Lasso |
0.080 |
0.278 |
0.827 |
Random Forest |
Random Forest |
Random Forest |
0.080 |
0.275 |
0.815 |