[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-665393-105":59,"doc-detail-665393-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","using-causal-diagrams-to-guide-analysis-in-missing-data-problems","Using Causal Diagrams to Guide Analysis in Missing Data Problems","","Estimating causal effects under incomplete observations requires additional, generally untestable assumptions about why data are missing. The paper shows how causal diagrams can express these assumptions, clarifying links between missingness classifications (MCAR, MAR, MNAR) and when unbiased causal effects can be obtained by analyzing only complete cases. It formally extends Pearl-style back-door reasoning to incomplete-data settings and illustrates the approach with an occupational cohort study of cosmic radiation exposure and skin cancer incidence.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/using-causal-diagrams-to-guide-analysis-in-missing-data-problems/665393/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/using-causal-diagrams-to-guide-analysis-in-missing-data-problems/665393.png","ImageObject",300,407,{"name":92,"@type":93},"miaandmiso","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-10-29",true,{"@type":102,"interactionType":103,"userInteractionCount":8},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why do missing-data problems require additional assumptions for causal estimation?","Question",{"text":112,"@type":113},"Causal effects from incomplete data rely on assumptions about the mechanism generating missingness. These assumptions are often inherently untestable, so causal validity depends on how close they are to the truth.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How do causal diagrams help in the missing-data context?",{"text":117,"@type":113},"Causal diagrams represent the additional assumptions about missingness mechanisms and clarify how these assumptions relate to classifications such as MCAR, MAR, and MNAR.",{"name":119,"@type":110,"acceptedAnswer":120},"When can causal effects be estimated without bias using only complete cases?",{"text":121,"@type":113},"The paper considers when the required assumptions are close to correct, allowing causal effects to be identified without bias from subjects with complete records. It also points readers to alternative methods when this is not justified.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},665393,1793302028,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":8,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":144},7971478544854,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","Using Causal Diagrams to Guide Analysis in Missing Data Problems  \nRhian M. Daniel∗, Michael G. Kenward, Simon N. Cousensand Bianca L. De Stavola  \nFaculty of Epidemiology and Population Health,  \nLondon School of Hygiene and Tropical Medicine  \nAbstract  \nEstimating causal effects from incomplete data requires additional and inherently untestable assumptions regarding the mechanism giving rise to the missing data. We show that using causal diagrams to represent these additional assumptions both complements and clarifies some of the central issues in missing data theory, such as Rubin’sclassification of missingness mechanisms (as missing completely at random (MCAR), missing at random (MAR) or missing not at random (MNAR)) and the circumstances in which causal effects can be estimated without bias by analysing only the subjects with complete data. In doing so, we formally extend the back-door criterion of Pearl and others for use in incomplete data examples. These ideas are illustrated with an example drawn from an occupational cohort study of the effect of cosmic radiation on skin cancer incidence.  \nKeywords: missing data; causal inference; causal diagram.  \n􀀃 Address for correspondence: Rhian M. Daniel, Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, Keppel Street, London  \nWC1E 7HT, UK. E-mail: [rhian.daniel@lshtm.ac.uk](rhian.daniel@lshtm.ac.uk)  \n1 Introduction  \nA key aim of medical and epidemiological research is to establish causal links between treatment, or other exposures, and outcomes. The gold-standard approach to achieve this aim is to conduct an ‘ideal’ randomised controlled trial (RCT), where by ‘ideal’we mean large, double-blind, with no missing data and full compliance. Such ideal RCTs ensure that the observed outcomes in different treatment arms are free from any systematic differences except for those induced by the treatments being compared.  \nAs we move from this ideal, causal inference increasingly requires further assumptions. Causal diagrams 1 ,2 can represent these and the accompanying theory is useful in informing the design and analysis of studies. Pearl and Greenland et al. 1 ,2 show how causal diagrams can be used to guide the choice of variables for data collection (and subsequent conditioning in the analysis), in order to make causal inferences more plausible.  \nCausal diagrams are increasingly used in non-randomised studies, where the main focus is on the control of naturally occurring confounding and investigator-induced selection bias. 3 The case for using causal diagrams to represent the mechanism assumed to give rise to missing data has not been extensively studied. In this article we fill this gap, showing that causal diagrams can complement and clarify some key issues relating to the analysis of incomplete data.  \nWe consider the problem of estimating the causal effect of exposure A (e.g. exposure to cosmic radiation) on outcome Y ([e.g.](e.g. skin)[ skin](e.g. skin) cancer), when either A, Y or both, are incompletely observed. We consider whether or not this causal effect can be estimated without bias using only the complete records, i. e. subjects for whom both exposure and outcome are observed, and refer readers to appropriate alternative methods when this is not the case. We often consider additional covariates Z, and consider estimating the causal effect of A on Y conditional on Z. In this case, the definition of a complete record becomes a subject for whom A, Y and Z are all observed.  \nThe assumptions underpinning a causal analysis can be divided in two:(a) causal assumptions, such as ‘missingness is affected by exposure to radiation’ and (b) parametric assumptions, such as ‘the logarithm of the odds of skin cancer increases linearly with age’. The causal effect of A on Y will only  \nbe estimated without bias if the assumptions made—both (a) and (b)—are close to being correct. We focus on assumptions of type (a) and assume that assumptions of type (","cbCaie8DYNV3owpM","https://ap.wps.com/l/cbCaie8DYNV3owpM","pdf",265783,46,"English","# Introduction\n## Motivating example: the British Commercial Airline Pilots and Air Traffic Control Officers Study\n### The data","[{\"question\":\"Why do missing-data problems require additional assumptions for causal estimation?\",\"answer\":\"Causal effects from incomplete data rely on assumptions about the mechanism generating missingness. These assumptions are often inherently untestable, so causal validity depends on how close they are to the truth.\"},{\"question\":\"How do causal diagrams help in the missing-data context?\",\"answer\":\"Causal diagrams represent the additional assumptions about missingness mechanisms and clarify how these assumptions relate to classifications such as MCAR, MAR, and MNAR.\"},{\"question\":\"When can causal effects be estimated without bias using only complete cases?\",\"answer\":\"The paper considers when the required assumptions are close to correct, allowing causal effects to be identified without bias from subjects with complete records. It also points readers to alternative methods when this is not justified.\"}]","Using Causal Diagrams to Guide Analysis in Missing Data Problems | PDF",116]