Dear friends in causality research,<\/p>\n
Welcome to the 2017 Mid-summer greeting from the Ucla Causality Blog.<\/p>\n
This greeting discusses the following topics:<\/p>\n
1. “The Eight Pillars of Causal Wisdom”\u00a0and the WCE 2017 Virtual Conference Website.
\n2. A discussion panel: “Advances in Deep Neural Networks”,
\n3. Comments on “The Tale Wagged by the DAG”,
\n4. A new book: “The book of Why”,
\n5. A new paper: Disjunctive Counterfactuals,
\n6. Causality in Education Award,
\n7. News on “Causal Inference: A \u00a0Primer”<\/p>\n
1. “The Eight Pillars of Causal Wisdom”<\/strong><\/p>\n The tenth annual West Coast Experiments Conference was held at UCLA on\u00a0April 24<\/span><\/span>-25, 2017, preceded by a training workshop \u00a0on\u00a0April 23<\/span><\/span><\/span>.<\/p>\n You will be pleased to know that the WCE 2017 Virtual Conference\u00a0Website is now available here: The conference brought together scholars and\u00a0graduate students in economics, political science and other\u00a0social sciences who share an interest in causal analysis. Speakers included:<\/p>\n 1. Angus Deaton, on Understanding and misunderstanding randomized\u00a0controlled trials. The video of my lecture “The Eight Pillars of Causal Wisdom”\u00a0can be watched here: 2. “Advances in Deep Neural Networks”<\/strong><\/p>\n As part of the its celebration of the 50 years of the Turing Award, the ACM has organized several discussion sessions on selected topics in computer science. I participated in a panel discussion on 3. The Tale Wagged by the DAG<\/strong><\/p>\n An article by this title, authored by Nancy Krieger and George\u00a0Davey Smith has appeared in the International Journal of Epidemiology, IJE\u00a02016 45(6) 1787-1808<\/span>. As the title tell-tales us, the authors are unhappy with the direction that modern epidemiology has taken, which is too wedded to a two-language framework: The specific reasons for the authors unhappiness\u00a0are still puzzling to me, because the article does not demonstrate concrete alternatives to current methodologies. I can only speculate however that it is\u00a0the dazzling speed with which epidemiology has modernized its tools that lies behind the authors discomfort. If so, it would be safe for us to assume that\u00a0the discomfort will subside as soon as researchers gain greater familiarity with the capabilities and flexibility of these new tools. \u00a0I nevertheless recommend that the article,\u00a0and the entire special issue of IJE be studied by our readers,\u00a0because they reflect an interesting soul-searching attempt by a forward-looking discipline to assess its progress in the wake of a profound paradigm shift.<\/p>\n Epidemiology, as I have written on several occasions, has been a pioneer in accepting the DAG-counterfactuals symbiosis as a ruling paradigm — way ahead of mainstream statistics and its other satellites. (The social sciences, for example, are almost there, with the exception of the\u00a0model-blind branch of econometrics. See Feb. 22 2017 posting)<\/p>\n In examining the specific limitations that Krieger and Davey Smith perceive in DAGs, readers will be amused to note that these limitations coincide precisely with the strengths\u00a0for which DAGs are praised.<\/p>\n For example, the article complains that\u00a0DAGs provide no information about variables that investigators chose not to include\u00a0in the model. \u00a0In their words:\u00a0“the DAG does not provide a comprehensive picture. For example, it does not include paternal factors, ethnicity, respiratory infections or socioeconomic position…” (taken from the Editorial introduction). I have never considered this to be a limitation of DAGs or of any other scientific modelling. Quite the contrary.\u00a0It would be a disaster if models were permitted to provide information unintended by the modeller. Instead, I have learned to admire the ease with which\u00a0DAGs enable researchers to incorporate knowledge about new variables, or new mechanisms, which the modeller wishes Model misspecification, after all, \u00a0is a problem that plagues every \u00a0exercise in causal inference,\u00a0no matter what framework one chooses to\u00a0adapt. It can only be cured by careful model-building Or, to take another example, the authors call repeatedly\u00a0for an ostensibly unavailable methodology which they label “causal triangulation” (it appears 19 times in the article). In their words: “In our field, involving dynamic populations of people in dynamic societies and ecosystems, methodical triangulation of diverse types of evidence from diverse types of study settings and involving diverse populations is essential.” \u00a0Ironically, however,\u00a0the task of treating “diverse type of evidence from\u00a0diverse populations” has been accomplished quite successfully in the dag-counterfactual framework. See, for example the formal and complete results of (Bareinbaum and Pearl, 2016,\u00a0http:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r450-reprint.pdf<\/a><\/span><\/span>) which have emerged from DAG-based perspective and invoke the do-calculus. (See also\u00a0\u00a0http:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r400.pdf<\/a>)\u00a0<\/span><\/span><\/span>It is inconceivable for me to imagine anyone\u00a0pooling data from two different designs (say Another conceptual paradigm which the authors hope would liberate us from the tyranny of DAGs and counterfactuals is Lipton’s (2004) romantic aspiration for “Inference to the Best Explanation.” It is a compelling, century old mantra, going back at least to Charles Pierce\u00a0theory of abduction (Pragmatism and Pragmaticism, 1870) which,\u00a0unfortunately, has never operationalized its key terms: “explanation,” “Best” and “inference to”. \u00a0Again, I know of only one framework in which this aspiration has\u00a0been explicated with sufficient precision to produce tangible\u00a0results — it is the structural framework of DAGs and counterfactuals. See, for example, Causes of Effects and Effects of Causes” In summary, what Krieger and Davey Smith aspire to achieve by abandoning the structural framework has already been accomplished with the help and grace of that very framework. I was pleased to note that, by and large, commentators on Krieger and Davey Smith paper seemed to be aware of the powers\u00a0and generality of the DAG-counterfactual framework,\u00a0albeit not exactly for the reasons that I have described here. [footnote: I have many disagreements with the other commentators as well, but I wish to focus here on the TALE WAGGED DAG where the problems appear more glaring.] My talk on “The Eight Pillars of Causal Wisdom”\u00a0provides a concise summary of those reasons\u00a0and explains why I take the poetic liberty of calling\u00a0these pillars “The Causal Revolution” All in all, I believe that epidemiologists should be commended for the incredible progress they have made in the past two decades. They will no doubt continue to develop and benefit from the new tools that the DAG-counterfactual symbiosis has spawn. At the same time, I hope that the discomfort that Krieger and Davey Smith’s have expressed will\u00a0be temporary and that it will inspire a greater understanding\u00a0of the modern tools of causal inference.<\/p>\n Comments on this special issue of IJE are invited on this blog.<\/p>\n 4. The Book of WHY<\/strong><\/p>\n As some of you know, I am co-authoring another book, titled: “The Book of Why: The new science of cause and effect”. It will attempt to present the eight pillars of causal wisdom to the general public using words, intuition and examples to replace equations. My co-author is science writer Dana MacKenzie (danamackenzie.com) and our publishing house is Basic Books. If all goes well, the book will see your shelf by\u00a0March 2018<\/span><\/span><\/span>. Selected sections will appear periodically on this blog.<\/p>\n 5. Disjunctive Counterfactuals<\/strong><\/p>\n The structural interpretation of counterfactuals as formulated in Balke and Pearl (1994) excludes \u00a0disjunctive conditionals, such as “had X been x1 or x2”, as well as disjunctive actions such as do(X=x1 or X=x2).\u00a0\u00a0In contrast, the closest-world interpretation of Lewis ( 1973) assigns truth values to\u00a0all counterfactual sentences, regardless of the\u00a0logical form of the antecedant.\u00a0The next issue of the Journal of Causal Inference will include a paper that extends the vocabulary of structural counterfactuals with disjunctions, and clarifies the assumptions needed for the extension. An advance copy can be viewed here: 6. \u00a0ASA Causality in Statistics Education Award<\/strong><\/p>\n Congratulations go to Ilya Shpitser,\u00a0Professor of Computer Science at Johns Hopkins University, who is the 2017 recipient of the ASA Causality in Statistics Education\u00a0Award. \u00a0Funded by Microsoft Research and Google, the $5,000 Award, will be presented to Shpitser at the 2017 Joint Statistical Meetings (JSM 2017) in Baltimore.<\/p>\n Professor Shpitser has developed Masters\u00a0level graduate course material that takes causal inference from the\u00a0ivory towers of research to the level of students with a machine\u00a0learning and data science background. It combines techniques of\u00a0graphical and counterfactual models and provides both an accessible\u00a0coverage of the field and excellent conceptual, computational and\u00a0project-oriented exercises for students.<\/p>\n These winning materials and those of the previous Causality in\u00a0Statistics Education Award winners are available to download online\u00a0at\u00a0http:\/\/www.amstat.org\/education\/causalityprize\/<\/a><\/span><\/span><\/span><\/p>\n Information concerning nominations, criteria and previous winners can be viewed here: 7. News on “Causal Inference: A Primer”<\/strong><\/p>\n Wiley, the publisher of our latest book “Causal Inference in\u00a0Statistics: A Primer” (2016, Pearl, Glymour and Jewell) is informing us that the book is now in its 4th printing, corrected\u00a0for all the errors we (and others) caught since\u00a0the first publications. To buy a corrected copy, make sure you get the “4th “printing”. The trick is to look at the copyright page and make sure Happy Summer-end, and may all your causes Dear friends in causality research, Welcome to the 2017 Mid-summer greeting from the Ucla Causality Blog. This greeting discusses the following topics: 1. “The Eight Pillars of Causal Wisdom”\u00a0and the WCE 2017 Virtual Conference Website. 2. A discussion panel: “Advances in Deep Neural Networks”, 3. Comments on “The Tale Wagged by the DAG”, 4. A […]<\/p>\n","protected":false},"author":7,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6,11,14],"tags":[],"class_list":["post-1806","post","type-post","status-publish","format-standard","hentry","category-counterfactual","category-discussion","category-epidemiology"],"_links":{"self":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/1806","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/comments?post=1806"}],"version-history":[{"count":2,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/1806\/revisions"}],"predecessor-version":[{"id":1808,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/1806\/revisions\/1808"}],"wp:attachment":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/media?parent=1806"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/categories?post=1806"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/tags?post=1806"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\n
\nhttp:\/\/spp.ucr.edu\/wce2017\/<\/a><\/span><\/span>
\nIt provides videos of the talks as well as some of the papers and\u00a0presentations.<\/p>\n
\n2. Chris Auld, on the on-going confusion between regression vs.\u00a0structural equations in the econometric literature.
\n3. Clark Glymour, on Explanatory Research vs Confirmatory Research.
\n4. Elias Barenboim, on the solution to the External Validity problem.
\n5. Adam Glynn, on Front-door approaches to causal inference.
\n6. Karthika Mohan, on Missing Data from a causal modeling perspective.
\n7. Judea Pearl, on “The Eight Pillars of Causal Wisdom.”
\n8. Adnan Darwiche, on Model-based vs. Model-Blind Approaches\u00a0to Artificial Intelligence.
\n9. Niall Cardin, Causal inference for machine learning.
\n10. Karim Chalak, Measurement Error without Exclusion.
\n11. Ed Leamer, “Causality Complexities Example: Supply and Demand.
\n12. Rosa Matzkin, “Identification is simultaneous equation.
\n13 Rodrigo Pinto, Randomized Biased-controlled Trials.<\/p>\n
\nhttps:\/\/www.youtube.com\/watch?v=8nHVUFqI0zk<\/a>
\nA transcript of the talk can be found here:
\nhttp:\/\/spp.ucr.edu\/wce2017\/Papers\/eight_pillars_of.pdf<\/a><\/span><\/span><\/p>\n
\n
\n“Advances in Deep Neural Networks”, which gave me an opportunity to share thoughts on whether learning methods based solely on data fitting can ever achieve a human-level\u00a0intelligence. The discussion video can be viewed here:
\nhttps:\/\/www.youtube.com\/watch?v=mFYM9j8bGtg<\/a><\/span><\/span><\/span>
\nA position paper that defends these thoughts is available here:
\nweb.cs.ucla.edu\/~kaoru\/theoretical-impediments.pdf<\/a><\/p>\n
\n
\nhttps:\/\/academic.oup.com\/ije\/issue\/45\/6#250304-2617148<\/a><\/span><\/span>
\nIt is part of a special IJE issue on causal analysis which, for the reasons outlined below, should be of interest\u00a0to readers of this blog.<\/p>\n
\n(1) Graphical models (DAGs) — to express what we know, and
\n(2) Counterfactuals (or potential outcomes) — to express\u00a0what we wish to know.<\/p>\n
\nto embrace.<\/p>\n
\nstrategies, and by enhancing the modeller’s knowledge. Yet, when it comes to minimizing misspecification errors, DAGS have no match. The transparency with which DAGs display the causal assumptions in the model, and the ease\u00a0with which the DAG identifies the testable implications of\u00a0those assumptions are incomparable; these facilitate speedy\u00a0model diagnosis and repair with no match in sight.<\/p>\n
\nexperimental and observational) without resorting to DAGs or (equivalently) potential outcomes, I am open to learn.<\/p>\n
\nhttp:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r431-reprint.pdf<\/a><\/span><\/span>
\nand Halpern and Pearl (2005) “Causes and explanations: A structural-model approach”
\nhttp:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r266-part1.pdf<\/a><\/span><\/span><\/p>\n
\nMore generally, what we learn from these examples is that the DAG-counterfactual symbiosis is far from being a narrow “ONE approach to causal inference” which ” may\u00a0potentially lead to spurious causal inference”\u00a0(their words). It is in fact a broad and flexible framework\u00a0within which a plurality of tasks and aspirations can be formulated,\u00a0analyzed and implemented. The quest for metaphysical alternatives is not warranted.<\/p>\n
\nhttp:\/\/spp.ucr.edu\/wce2017\/Papers\/eight_pillars_of.pdf<\/a><\/span><\/span><\/p>\n
\n
\n
\nhttp:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r459.pdf<\/a><\/span><\/span><\/p>\n
\n
\nhttp:\/\/www.amstat.org\/ASA\/Your-Career\/Awards\/Causality-in-Statistics-Education-Award.aspx<\/a><\/span><\/span>
\nand here:
\nhttp:\/\/magazine.amstat.org\/blog\/2012\/11\/01\/pearl\/<\/a><\/span><\/span><\/p>\n
\n
\nthe last line reads:\u00a010 9 8 7 6 5 4
\n<\/span>
\nIf you already have a copy, look up our errata page,
\nhttp:\/\/web.cs.ucla.edu\/~kaoru\/BIB5\/pearl-etal-2016-primer-errata-pages-may2017.pdf<\/a><\/span><\/span>
\nwhere all corrections are marked in red. The publisher also tells us the the Kindle version is much improved. I hope you concur.<\/p>\n
\n
\nproduce healthy effects.
\nJudea<\/p>\n","protected":false},"excerpt":{"rendered":"