[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126379-en":3,"doc-seo-126379-105":31,"detail-sidebar-cat-0-en-105":93},{"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126379,962085564549,"Genevieve","https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","General targeted machine learning for modern causal mediation analysis - academic paper","Causal mediation analyses study how effects of exposures on outcomes operate through mechanisms involving intermediate variables, supporting advances across scientific fields. While non-parametric definitions and identification of mediational effects have progressed, estimation methods lag—especially for non-parametric settings with multiple continuous or high-dimensional mediators. This paper introduces an all-purpose one-step estimation algorithm that couples targeted learning with machine learning across six common mediation parameters, yielding √n-convergence and asymptotic normality, validated by simulation and illustrated on pain management and opioid use disorder.","General targeted machine learning for modern causal  \narXiv :2408 . 14620v2 [ stat .ML] 12 Jun 2025  \nmediation analysis  \nRichard Liu 1,* , Nicholas T. Williams2 , Kara E. Rudolph2 , and Ivn D´ıaz 1 1Division of Biostatistics, Department of Population Health, New York University Grossman School of  \nMedicine, USA  \n2Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY,  \nUSA.  \n*Corresponding Author. Email: [richard.l@nyu.edu](richard.l@nyu.edu)  \nJune 13, 2025  \nAbstract  \nCausal mediation analyses investigate the mechanisms through which causes exert their effects, and are therefore central to scientific progress. The literature on the non-parametric definition and identification of mediational effects in rigorous causal models has grown significantly in recent years, and there has been important progress to address challenges in the interpretation and identification of such effects. However, statistical methodology for nonparametric estimation has lagged, with few or no methods available for tackling non-parametric estimation in the presence of multiple, continuous, or high-dimensional mediators. In this paper we propose an all-purpose one-step estimation algorithm that can be coupled with machine learning in mediation studies that use any of six common mediation parameters (natural direct and indirect effects, randomized interventional effects, separable effects, organic direct and  \nindirect effects, recanting twin effects, and decision theoretic effects) . The estimators build on methods for double machine learning, including a re-parameterization of the identification formulas in terms of sequential regressions, a first order non-parametric von-Mises approximation of the first-order bias of a plug-in estimator, and Riesz learning for estimation of nuisance parameters. We show that the proposed one-step estimators have desirable properties, such as √n-convergence and asymptotic normality. We illustrate the properties of our methods ina simulation study and demonstrate its use on real data to estimate the extent to which pain management practices mediate the total effect of having a chronic pain disorder on opioid use disorder. We provide an R package (on CRAN) implementing our methods publicly available  \nat [https://github.com/nt-williams/crumble](https://github.com/nt-williams/crumble).  \n1 Introduction  \n1.1 Prior literature  \nCausal mediation analyses seek to investigate the extent through which the effect of an exposure on an outcome operates through effects of the exposure on intermediate variables. Recent decades have seen increased attention to the development of methodology for the definition, identification, and estimation of effects for mediation analysis, as well as their increased application in various scientific fields. Particularly, the definition and identification of effects that are guaranteed to measure the mechanisms through which the effect operates has been the subject of considerable debate and methodological work (e.g., Robins, 2003 ; Pearl, 2010 ; Robins and Richardson, 2010 ; Miles, 2023 ; D´ıaz, 2024) .  \nNatural direct and indirect effects (NDE and NIE; Robins and Greenland, 1992 ; Pearl, 2001) are one of the most widely known definitions of causal mediation parameters with an agreed upon mechanistic interpretation. Broadly, the NDE measures the effect that operates independent of a mediator through counterfactual variables in which the effect through the mediator is disabled, whereas the NIE measures effects through the mediator by considering counterfactuals that disable all effects that operate independently of it. In spite of their scientific importance, desirable interpretation, and widespread use, the NDE and NIE are seldom identified from observed data, because their identification relies on so-called cross-world counterfactual assumptions—independence assumptions between counterfactual variables indexed by distinct interventions on the exposu","cbCaikIB719lHeXT","https://ap.wps.com/l/cbCaikIB719lHeXT","pdf",625966,6,1,56,"English","en",105,"# Abstract\n# Introduction\n## Prior literature","[{\"question\":\"What problem does the paper address in causal mediation analysis?\",\"answer\":\"It addresses the lack of statistical methodology for non-parametric estimation when mediation includes multiple continuous or high-dimensional mediators.\"},{\"question\":\"How does the proposed method estimate mediation effects?\",\"answer\":\"It uses an all-purpose one-step estimation algorithm coupled with machine learning, building on double machine learning, sequential regressions, von-Mises bias approximation, and Riesz learning for nuisance parameters.\"},{\"question\":\"What theoretical and empirical results are provided?\",\"answer\":\"The one-step estimators are shown to have √n-convergence and asymptotic normality, with properties demonstrated through simulations and an application to real data on pain management practices and opioid use disorder.\"}]","General targeted machine learning for modern causal mediation analysis - academic paper | PDF",1785904758,141,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"general-targeted-machine-learning-for-modern-causal-mediation-analysis-academic-paper","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/general-targeted-machine-learning-for-modern-causal-mediation-analysis-academic-paper/126379/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What problem does the paper address in causal mediation analysis?","Question",{"text":77,"@type":78},"It addresses the lack of statistical methodology for non-parametric estimation when mediation includes multiple continuous or high-dimensional mediators.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How does the proposed method estimate mediation effects?",{"text":82,"@type":78},"It uses an all-purpose one-step estimation algorithm coupled with machine learning, building on double machine learning, sequential regressions, von-Mises bias approximation, and Riesz learning for nuisance parameters.",{"name":84,"@type":75,"acceptedAnswer":85},"What theoretical and empirical results are provided?",{"text":86,"@type":78},"The one-step estimators are shown to have √n-convergence and asymptotic normality, with properties demonstrated through simulations and an application to real data on pain management practices and opioid use disorder.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":108,"slug":139},19,"General","general"]