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
2.56k stars 444 forks source link

Issue 365: Regression of treatment on covariates #366

Closed akaashp2000 closed 6 months ago

akaashp2000 commented 7 months ago

Resolves https://github.com/matheusfacure/python-causality-handbook/issues/365

There are some lines in this chapter that caused me some confusion, but want to make sure others agree or can explain before I propose a change.

There are some lines on residual from prediction of $T$ (treatment) from the other features $X{1i}, ..., X{ki}$

a) "where $\tilde{Ti}$ is the residual from a regression of all other covariates $X{1i} + ... + X_{ki}$ on $Ti$" (under Regression Theory). Should this mean regression of $T$ on $X{1}, ..., X_{k}$ - not a sum? Since treatment is being predicted from the covariates?

b) "where $\delta_{A}$ is the vector of coefficients from the regression of $A$ on $Educ$" (under Regression for Non-Random Data)? Again should this be regression of $Educ$ on $A$?

Thanks!