[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119878-en":3,"doc-seo-119878-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119878,1649267921044,"Ava Thompson","https://us-avatar.wpscdn.com/avatar/1800007509477c92dfb?_k=1782875107921204101",8,"Research & Report","LONG STORY SHORT - OMITTED VARIABLE BIAS IN CAUSAL MACHINE LEARNING","We derive general, sharp bounds on the magnitude of omitted variable bias for a wide class of common causal parameters, covering weighted averages of potential outcomes, average treatment effects (including subgroup effects such as the effect on the treated), average causal derivatives, and policy effects induced by covariate distribution shifts. The bounds are built for general semiparametric and fully nonparametric regression models using the Riesz-Frechet representation, so the bias depends only on variation created by latent variables in both the outcome regression and the causal parameter’s Riesz representer.","arXiv :2112 . 13398v4 [ econ .EM] 2 Nov 2023  \nLONG STORY SHORT: OMITTED VARIABLE BIAS IN CAUSAL MACHINE  \nLEARNING  \nVICTOR CHERNOZHUKOV†, CARLOS CINELLI* , WHITNEY NEWEY‡, AMIT SHARMA∥ ,  \nAND VASILIS SYRGKANIS§  \nABSTRACT. We derive general bounds on the size of omitted variable bias for a broad class of common causal parameters, such as (weighted) average of potential outcomes, average treatment effects (including subgroup effects, such as the effect on the treated), average causal derivatives, and policy effects from shifts in covariate distribution—all for general, semiparametric and fully nonparametric regression models. Leveraging the Riesz-Frechet representation of the target parameter, we show that the bounds on the bias depend only on the additional variation that latent variables create both in the outcome regression and in the Riesz representer of the causal parameter of interest.  \nWe further show how simple plausibility judgments on the maximum explanatory power of latent variables are sufficient to place overall bounds on the size of the bias. Finally, to take the bounds to data, we develop flexible and efficient statistical inference methods on the learnable components of the bounds, which can make use of modern machine learning algorithms for estimation. These results allow empirical researchers to perform sensitivity analyses in a flexible class of machine-learned causal models using very simple, and interpretable, tools. We demonstrate the usefulness of the approach with two empirical examples.  \nKeywords: sensitivity analysis, short regression, long regression, omitted variable bias, omitted confounders, causal models, machine learning, confidence bounds.  \n† Dept. of Economics, Massachusetts Institute of Technology, Cambridge, MA, USA. Email: [vchern@mit.edu](vchern@mit.edu).  \n* Dept. of Statistics, University of Washington, Seattle, WA, USA. Email: [cinelli@uw.edu](cinelli@uw.edu).  \n‡ Dept. of Economics, Massachusetts Institute of Technology, Cambridge, MA, [USA. Email: wnewey@mit.edu](USA. Email: wnewey@mit.edu).∥ Microsoft Research India, Bangalore, India. Email: [amshar@microsoft.com](amshar@microsoft.com) .  \n§ Dept. of Mgmt Science and Engineering, Stanford University, Stanford, CA, USA. Email: [vsyrgk@stanford.edu](vsyrgk@stanford.edu). Date: November 3, 2023 . First ArXiv version: December 2021 .  \nThis is an extended version of an earlier paper prepared for the NeurIPS-21 Workshop “Causal Inference & Machine Learning: Why now?”. We thank Elias Bareinboim, Ben Deaner, David Green, Judith Lok, Esfandiar Maasoumi, Steve Lehrer, Richard Nickl, Jack Porter, James Poterba, Eric Tchetgen Tchetgen, Ingrid Van Keilegom, and also participants of the Chambelain seminar, Canadian Economic Association the Institute for Nonparametric, and Uncertainty in Artificial Intelligence meetings, and seminars at Harvard-MIT, Wisconsin, Emory, and BU Causal Seminar for very helpful comments. We are grateful to Jack Porter for suggesting the long story short title.  \n1  \n1. INTRODUCTION  \nCausal inference with observational data usually relies on the assumption that the treatment assignment mechanism is “ignorable”(i.e, independent of potential outcomes) conditional on a set of observed variables; or, equivalently, that the set of observed covariates satisfy the “backdoor”(or, more generally, adjustment) criterion (Rosenbaum and Rubin, 1983a; Pearl, 2009; Angristand Pischke, 2009; Shpitser et al., 2012; Imbens and Rubin, 2015) . Investigators who rely on the conditional ignorability assumption for drawing causal inferences from non-experimental studies must, therefore, also be able to cogently argue that there are no unobserved confounders of the treatment-outcome relationship. Yet, claiming the absence of unmeasured confounders is not only fundamentally unverifiable from the data, but often an assumption that is very hard to defend in practice.  \nWhen the assumption of no unobserved confounders is called into question,","cbCaitdedQ9RjKuT","https://ap.wps.com/l/cbCaitdedQ9RjKuT","pdf",707691,1,55,"English","en",105,"# Abstract\n# Introduction\n## Sensitivity analysis under unobserved confounding\n## Limitations of prior sensitivity methods\n## Main contribution: sharp bounds via linear functionals","[{\"question\":\"What does the paper bound about omitted variable bias?\",\"answer\":\"It derives general, sharp bounds on the size of omitted variable bias for a broad class of causal parameters, including average treatment effects, derivatives, and policy effects.\"},{\"question\":\"How are the bounds constructed for different regression models?\",\"answer\":\"The results hold for general semiparametric and fully nonparametric regression models, relying on the Riesz-Frechet representation of the target parameter.\"},{\"question\":\"How can researchers use these results in practice?\",\"answer\":\"The paper explains that simple plausibility judgments about the maximum explanatory power of latent variables can place overall bounds on bias, and it develops efficient inference methods to connect the bounds to data.\"}]","LONG STORY SHORT - OMITTED VARIABLE BIAS IN CAUSAL MACHINE LEARNING | PDF",1785726800,139,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"long-story-short-omitted-variable-bias-in-causal-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/long-story-short-omitted-variable-bias-in-causal-machine-learning/119878/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What does the paper bound about omitted variable bias?","Question",{"text":75,"@type":76},"It derives general, sharp bounds on the size of omitted variable bias for a broad class of causal parameters, including average treatment effects, derivatives, and policy effects.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How are the bounds constructed for different regression models?",{"text":80,"@type":76},"The results hold for general semiparametric and fully nonparametric regression models, relying on the Riesz-Frechet representation of the target parameter.",{"name":82,"@type":73,"acceptedAnswer":83},"How can researchers use these results in practice?",{"text":84,"@type":76},"The paper explains that simple plausibility judgments about the maximum explanatory power of latent variables can place overall bounds on bias, and it develops efficient inference methods to connect the bounds to data.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]