We just released a new version (0.7.1) of the Modified Causal Forest (mcf). The mcf is a Python Causal Machine Learning package for estimating effects at all aggregation levels as well as performing optimal policy analysis. It is freely available on PyPI (pip install mcf).
Michael Lechner
@MichaelLechner8
Professor of Econometrics at the University of St. Gallen. Research interests in Causal Machine Learning, labour and sports economics.
240 Following 472 Followers 657 Posts
Latest posts 5
Our new paper (jointly with Jana Mareckova) on Comprehensive Causal Machine Learning (arxiv.org/abs/2405.10198) compares the properties and performance of Double Machine Learning, the Generalized Causal Forest and the Modified Causal Forest. Interesting differences appear!
In case you are interested in a non-technical introduction into Causal Machine Learning for Economists, SJES just published my paper (sjes.springeropen.com/articles/10.11…). It's open-access and based on a keynote I gave at the meeting of the German Economic Association recently.
After some longer journey our paper analysing the Flemish Active Labour Market Policy by Causal Machine Learning methods finally found a nice home. authors.elsevier.com/c/1gDYk3IvSGh4…
Wonderful!