Why we can use E[T|X, W, Z] as the instrument in DRIV/OrthoIV approaches?
Nobody has claimed this yet.
- Dominant language
- Jupyter Notebook
- Stars
- 4.8k
- Forks
- 827
- PR merge metrics
- No merged PRs in 30d
Description
I am not sure why it is reasonable to use E[T|X, W, Z] as the Z in DRIV/OrthoIV. Is there any paper I can refer to? Also, In my simulation, it seems that we need to perform this projection when the instrument is weak (i.e., a small correlation between T and Z). I wonder whether the weak instrument case is the case for which the projection approach is designed?
Thanks!
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
No file, test, or entry point is named in the issue. Research the assumptions behind DRIV/OrthoIV and the projection E[T|X, W, Z], then provide paper references and explain whether the projection is specifically motivated by weak instruments; define done as a clear, supported answer to both questions.
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
- 15/100