Binary Treatment & Outcome. The path to the right solution
Nobody has claimed this yet.
- Dominant language
- Jupyter Notebook
- Stars
- 4.8k
- Forks
- 827
- PR merge metrics
- No merged PRs in 30d
Description
Hello! First of all, thank you for this amazing library, it's gorgeous.
But in the process of research, I had several questions about the methodology and interpretation.
Question 1: we have the "effect" method in CausalForest - using it I get effect estimates greater than 1, while my outcome and treatment are binary (0 and 1 only). The actual question is, are the values in the range [-1; 1] a percentage of the effect (for example, 0.12 = 12% is the effect relative), and the values outside the interval are the effect absolute (12 = 12% is the effect absolute)? And what is the ATE score in this case?
Question 2: Methodologically, reading the threads, I saw that it is possible to model binary outcome and treatment in causalForest using regressor for Y and classifier for T. Is it so? I would appreciate it if you could explain this a little bit.
Have a nice day!
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the CausalForest.effect entry point and the ATE interpretation, then review the supported modeling choices for binary outcomes and treatments. Done means documenting how effect values and ATE should be interpreted and clarifying whether regressor and classifier models are valid for the stated inputs.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100