Chapter 11&21 Should ipw be used for xlearner?

matheusfacure / python-causality-handbook

Causal Inference for the Brave and True. A light-hearted yet rigorous approach to learning about impact estimation and causality.

https://matheusfacure.github.io/python-causality-handbook/landing-page.html

MIT License

2.65k stars 463 forks source link

Chapter 11&21 Should ipw be used for xlearner? #403

Open 21Timothy opened 1 month ago

21Timothy commented 1 month ago

I'm wondering do I need to use ipw for xlearner, as the last step of xlearner has used ps value to re-weight sample? Need help for this question, thanks!

21Timothy commented 1 month ago

And do I need to sampling for calculating ps value? When I reading the chapter 21, it seems that there's not need to do so, as the g(x) has the information of how unbalance it is of the train data?