Causal Inference Book

There&#x27;s also this primer, recently out: <a href="https:&#x2F;&#x2F;www.amazon.com&#x2F;Causal-Inference-Statistics-Judea-Pearl&#x2F;dp&#x2F;1119186846" rel="nofollow">https:&#x2F;&#x2F;www.amazon.com&#x2F;Causal-Inference-Statistics-Judea-Pea...</a>

There are very interesting and fairly recent results on causal discovery under additive noise models [1]. Although such models aren&#x27;t universally applicable the underlying concept is intuitive and a good fit for many problem domains. On the time series front, google open sourced CausalImpact [2] for Bayesian structural time-series modeling a few years ago. Looks like RankScience [3] is putting that research to good use.<p>I&#x27;m surprised that causal discovery doesn&#x27;t get more play. Aside from the direct applications to scientific research, causality is a strong consistency hint for assessing models&#x2F;features learned from data.<p>[1] <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=8776582" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=8776582</a><p>[2] <a href="https:&#x2F;&#x2F;opensource.googleblog.com&#x2F;2014&#x2F;09&#x2F;causalimpact-new-open-source-package.html" rel="nofollow">https:&#x2F;&#x2F;opensource.googleblog.com&#x2F;2014&#x2F;09&#x2F;causalimpact-new-o...</a><p>[3] <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=13552862" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=13552862</a>

Check this one out, it is the classic in book on causal inference: <a href="https:&#x2F;&#x2F;www.amazon.com&#x2F;Experimental-Quasi-Experimental-Designs-Generalized-Inference&#x2F;dp&#x2F;0395615569" rel="nofollow">https:&#x2F;&#x2F;www.amazon.com&#x2F;Experimental-Quasi-Experimental-Desig...</a>

It makes me feel sorry for the researchers that Harvard puts ads at the bottom of their web page.

I read through about 10 pages of the first book (&quot;without models&quot;). It struck me as very nearly identical to current statistics practice. It clearly differentiated itself as discussing counterfactuals (the data needed to actually determine causality) but I could not find the section of the book that described how counterfactual data can be inferred from missing data (without it being &quot;turtles all the way down&quot;).<p>Does anyone in this area have a succinct way to explain how counterfactual data can be inferred by these techniques - and how traditional statistics practice is not able to perform this inference?

Haven&#x27;t read the book, but the concept of causal inference from event data really deserves more attention. Automatic&#x2F;assisted cause analysis in complex systems has huge potential.

I have seen an earlier version of this book recommended here on HN, and thought that some might be delighted to know that there is a revised version available to download.

Surely the title should be &quot;causal&quot;?