py-why / py-why/EconML

Can we compare performance of DML estimators to DR Estimators based on the output of score method?

Open
#556 1 comment 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
4.8k
Forks
827
PR merge metrics
No merged PRs in 30d

Description

Hi,
Thanks for the great package and super helpful documentation.
I am working on a problem for estimating HTE's for a binary treatment and tried a couple of DML as well as DR estimators. I noticed that that DR estimators score on a validation set are 5 times greater than that of DML estimator scores .
Does that mean DML estimators are relatively better for my problem and why? Or output of score() in DML or DR can only be used to compare models within the respective classes? Also, what's the best metric to use for model selection? for e.g. Rscorer()

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the score() documentation for LinearDRLearner and LinearDML, then read the RScorer documentation and the surrounding estimator evaluation guidance. Done means clarifying whether these scores are comparable across estimator classes and identifying the documented metric or procedure recommended for model selection.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
analytics, machine-learning
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
28/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.