py-why / EconML

ALICE (Automated Learning and Intelligence for Causation and Economics) is a Microsoft Research project aimed at applying Artificial Intelligence concepts to economic decision making. One of its goals is to build a toolkit that combines state-of-the-art machine learning techniques with econometrics in order to bring automation to complex causal inference problems. To date, the ALICE Python SDK (econml) implements orthogonal machine learning algorithms such as the double machine learning work of Chernozhukov et al. This toolkit is designed to measure the causal effect of some treatment variable(s) t on an outcome variable y, controlling for a set of features x.

https://www.microsoft.com/en-us/research/project/alice/

Other

3.82k stars 714 forks source link

Issue with #791: Missing data allowed when training, but not inferencing #827

Open esbraun opened 12 months ago

esbraun commented 12 months ago

791 ought to allow missing data when inferencing if allow_missing = True, but currently an error is thrown when the inference data includes missing data. RE: below for a working example based on the "Metalearner Examples" notebook for a demonstration of the issue.

https://colab.research.google.com/drive/1QT-IdwRWRUiQNDwqJ0EXPXSftDuIg3uE?usp=sharing

Ftrejo23 commented 11 months ago

I'm seeing the same issue

fverac commented 11 months ago

Thanks for raising this issue. In the original PR we originally only considered missing data during training, but it seems there is interest in extending this to inferencing as well.

erasedcitizen11 commented 3 months ago

Hi there, just wondering if there any update on this as running into the same problems (doesn't allow missing values when computing CATE/ATE).

esbraun commented 3 months ago

@erasedcitizen11 I created a fork that disables the missing value checks which is a requirement for our use cases. Use at your own risk: https://github.com/esbraun/EconML/tree/main