Since @Andrew___Baker called for a break day, let's go back to our favorite Twitter activity of 2020... discussing DAGs! I'm very happy that our paper "Causal Inference and Data Fusion in Econometrics" is finally forthcoming in the Econometrics Journal. academic.oup.com/ectj/advance-a… 1/ 
In this paper, we review the advances that have been made in the causal AI literature in recent years and discuss their value for empirical work in econometrics and adjacent disciplines (such as political science, sociology, and management). 2/
We're not the first to discuss DAGs from an econometric perspective. Several famous scholars, including Jim Heckman, Hal White, and Dan McFadden were engaging with the topic before. Perhaps most notably, Guido Imbens published his comparison of .. 3/ aeaweb.org/articles?id=10…
.. DAGs and the potential outcomes framework in JEL in 2020. I wrote an entire series of Twitter threads, in which I discuss the points made in the paper and explain where I disagree, back then. You can find them searching for the hashtag #EconBookClub.
https://twitter.com/PHuenermund/status/1152169284677844993?s=204/
What was clear to us though is that none of these papers actually capture the current state of the art in the literature. So much has happened since 2009, when the 2nd ed. of Causality by @yudapearl, that we thought a survey of the latest advances would provide a lot of value. 5/
With the huge publication lags in econ it's actually quite hard to produce an up-to-date review of the fast-moving AI literature. And probably we could already start writing on a part 2 by now — which we actually should do at some point... 6/
But the paper will hopefully give you a good overview of the fundamental mechanics of the graphical framework to causality and an impression of how powerful it is in tackling some of the most thorny problems in applied empirical work. 7/
If you've seen an earlier version of the paper, you might want to have a second look. During the revision process, we've added several new examples for the backdoor and frontdoor criteria, as well as recovering from selection bias and identification by surrogate experiments. 8/
The paper will appear as part of a special issue on the works of Philip G. Wright, the inventor of IV and father of Sewall Wright, who developed path diagrams — an early version of DAGs. I'm happy that it will be joined by several fantastic contributions from other colleagues. 9/
The preprint version of the paper is available on arXiv: arxiv.org/abs/1912.09104 Let us know what you think. We're curious to hear your feedback! 🙏 10/10
• • •
Missing some Tweet in this thread? You can try to
force a refresh
