search

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
3.75k stars 711 forks source link

Question about data generating process in experiments #748

Open nehargupta opened 1 year ago

nehargupta commented 1 year ago

Hi,

For the experiment provided for the metalearners in this file https://github.com/py-why/EconML/blob/main/notebooks/Metalearners%20Examples.ipynb

in section 1.1, "We use the data generating process (DGP) from Kunzel et al."

Hmm....is that true? That's not the data generating process I think I'm seeing...is it possible you're referring to an old version of the paper or have adjusted the experiment somehow? Or, if I'm wrong, would be a huge help if you can refer me to the right section of the paper. Thanks!

kbattocchi commented 1 year ago

Hi, I believe what we meant is that the high-level structure of the DGP is taken from the paper, but you're right that the specific instantiations of e(x), mu_0(x), and mu_1(x) don't correspond exactly to any of the specific definitions from the appendix of that paper (though the treatment effect here is the same as that from their first simulation).