python-causality-handbook
python-causality-handbook copied to clipboard
matheusfacure
Causal Inference for the Brave and True. A light-hearted yet rigorous approach to learning about impact estimation and causality.
In the LATE section, > " We also rewrite potential outcomes with a double indexing notation, where the first index denotes the counterfactual for the instrument and the second index,...
"Bad COP" is an excellent part. But could you clarify what is the "COP" by definition or intuition? After reading the "Bad COP", I realized it is a backdoor to...
Hi everyone I would like to propose an alternative explanation about the "Randomized Experiments", especially in "potential outcomes are independent of the treatment". The current explanation is still confusing, but...
'...but only lower prices for the A population.' I guess it should be B not A.
There is an issue on chapter 25, in the following chart: It The labels are flipped of California vs. control. Really nice nootbook!
in the push example, the first non-IV regression controlled for push_assigned which is wrong if we are interested in the effect of push_delivered.
A chapter on doubly robust TMLE with python implementation would be super usefull !!
First of all, thank you much for the awesome resource! Whilst reading it, I noticed a few typos and a few sentences that didn't flow particularly well. So, please excuse...
In the section 20-Plug-and-Play Estimators under the part "Target Transformation", you have the following: "Also, we know that $Y_i * T_i = Y(1)_i*T_i$ and $Y_i * (1 - T_i )=...
There is an issue on chapter [18 - Heterogeneous Treatment Effects and Personalization](https://matheusfacure.github.io/python-causality-handbook/18-Heterogeneous-Treatment-Effects-and-Personalization.html), in the following paragraph [Predicting Sensitivity](https://matheusfacure.github.io/python-causality-handbook/18-Heterogeneous-Treatment-Effects-and-Personalization.html#predicting-sensitivity) > Here is an idea. What if we use linear regression?...
Metadata
Owner
Metadata
Causal Inference for the Brave and True. A light-hearted yet rigorous approach to learning about impact estimation and causality.