Timeline for Doubly robust learning with binary treatment and outcome
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 30, 2022 at 22:04 | comment | added | jkortner | How would you recommend to bound the output? @Lars. | |
| Sep 24, 2021 at 19:34 | comment | added | Lars | The function you used is fine. You can just bound the outputted predictions into [-1,1] manually. You might not want to use xgboost/random-forests/tree-based algorithms since they will lead to bound violations more often than smoother/spline-based estimators. | |
| Sep 23, 2021 at 8:43 | comment | added | bjarkemoensted | Thanks. Do you know if there's a generally recommended method of estimating binary treatment effects then? Of course I could train a classifier and observe the effect of the treatment parameter, but that will suffer from regularization bias, so I'm not sure if that's the best method available. | |
| S Sep 22, 2021 at 17:28 | review | First answers | |||
| Sep 22, 2021 at 17:37 | |||||
| S Sep 22, 2021 at 17:28 | history | answered | Lars | CC BY-SA 4.0 |