[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128197-en":3,"doc-seo-128197-105":31,"detail-sidebar-cat-0-en-105":92},{"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},128197,2336475104042,"Skyler","https://ap-avatar.wpscdn.com/avatar/22000c4c32af1715be0?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786537525561427321",8,"Research & Report","Causal Inference with Machine Learning for Complex Data Contexts","As data volume, setting complexity, and computation increase, causal inference research must produce robust, efficient methods for more intricate counterfactual questions. This dissertation develops machine-learning and semiparametric approaches for extracting and leveraging treatment-effect information in complex observation structures. It presents semiparametric estimation for hybrid trials with external controls, hierarchical modeling using randomization to reveal latent treatment-effect information, and proxy potential outcomes to improve efficiency. It also extends active feature acquisition for costly features in precision medicine and generalizes AFA with a nonparametric acquisition-conditioned oracle balancing acquisition cost and decision performance.","CAUSAL INFERENCE WITH MACHINE LEARNING FOR COMPLEX DATA CONTEXTS  \nMichael Valancius  \nA dissertation submitted to the faculty at the University of North Carolina at Chapel Hill in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biostatistics in the Gillings School of Global Public Health.  \nChapel Hill  \n2024  \nApproved by:  \nMichael R. Kosorok  \nMichael Hudgens Didong Li  \nDonglin Zeng  \nJunier Oliva  \n©2024 Michael Valancius ALL RIGHTS RESERVED  \nii  \nABSTRACT  \nMichael Valancius: Causal Inference with Machine Learning for Complex Data Contexts  \n(Under the direction of Michael R. Kosorok)  \nAs data volume, setting complexity, and computational capabilities expand, researchers in causal inference are tasked with developing methods that can meet the demands of this evolving landscape while maintaining robustness, efficiency, and (ultimately) effectiveness in addressing increasingly intricate counterfactual questions. This dissertation addresses challenges in causal inference within complex data contexts, where the nuances of the observation patterns and the data’s structure open both theoretical challenges and pathways for methodological innovation. Drawing on recent advancements in machine learning for causal inference and semiparametric theory, this dissertation develops new methodologies that aim to more fully leverage available information about treatment effects across several diverse settings. Chapters 2 and 3 focus on estimating causal effects in hybrid trials with external controls, systematically addressing the tradeoff between causal assumptions and efficiency bounds through a semiparametric lens. Chapter 4 explores hierarchical data structures, where a specific treatment randomization scheme uncovers previously unobserved information about treatment effects. In this context,‘proxy potential outcomes’ are introduced as a novel tool for approximating the joint distribution of potential outcomes to improve estimation efficiency. In precision medicine (Chapter 5) contexts where acquiring certain features is costly, we extend the active feature acquisition (AFA) framework to prioritize gathering only the most relevant information, minimizing acquisition costs while optimizing treatment recommendations. Chapter 6 further develops AFA methodology in a more general context, where multiple feature sets can be acquired sequentially. Here, we introduce anonparametric Acquisition Conditioned Oracle (ACO), which balances acquisition costs with prediction and decision-making objectives. Collectively, these contributions underscore the im-  \nportance of critically examining the origins and structure of treatment effect information within complex data contexts to enhance the rigor and efficiency of causal inference.  \nACKNOWLEDGEMENTS  \nThe true reward of completing a dissertation lies not in its conclusion but in the discoveries made along the way—about the subjects we study and, if we are fortunate, about ourselves. On this path, I have been deeply humbled by the kindness of those who have generously shared their knowledge, encouragement, and support. Their influence has shaped my abilities as a researcher, enriched my understanding of community, and guided my personal growth. To those who have shared their time and offered their guidance, I extend my heartfelt gratitude.  \nMy interest in statistics began as a teenager who was captivated by how the analytical revolution in baseball was driving the Tampa Bay Rays from the basement of Major League Baseball to the World Series. To my parents, Michael and Elizabeth, thank you sincerely for encouraging me to pursue my passions. Your support and willingness to let me explore my interests sparked a lifelong curiosity that has been invaluable in my research journey. Whether driving me to baseball games or listening to my endless baseball analysis, your support in my early statistical pursuits enabled me to pursue this path.  \nI would ","cbCaial3BQ0u15jS","https://ap.wps.com/l/cbCaial3BQ0u15jS","pdf",2665760,3,1,198,"English","en",105,"# Abstract\n# Acknowledgements\n## Motivation and research interests\n## Academic and advisor support\n## Committee and collaboration","[{\"question\":\"What problem does the dissertation focus on?\",\"answer\":\"It addresses how to develop causal inference methods that remain robust and efficient when data settings are complex and counterfactual questions become more intricate.\"},{\"question\":\"How do Chapters 2 and 3 handle causal effect estimation?\",\"answer\":\"They focus on estimating causal effects in hybrid trials with external controls, explicitly addressing the tradeoff between causal assumptions and efficiency bounds via semiparametric methods.\"},{\"question\":\"What role do active feature acquisition methods play?\",\"answer\":\"In precision medicine contexts with costly features, the dissertation extends active feature acquisition to gather only the most relevant information, and further generalizes it to sequential acquisition using an acquisition-conditioned oracle.\"}]","Causal Inference with Machine Learning for Complex Data Contexts | PDF",1785945495,499,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":29},"causal-inference-with-machine-learning-for-complex-data-contexts","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,51],{"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":20},"https://docshare.wps.com/document/research-report/",{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/causal-inference-with-machine-learning-for-complex-data-contexts/128197/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-27","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the dissertation focus on?","Question",{"text":76,"@type":77},"It addresses how to develop causal inference methods that remain robust and efficient when data settings are complex and counterfactual questions become more intricate.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How do Chapters 2 and 3 handle causal effect estimation?",{"text":81,"@type":77},"They focus on estimating causal effects in hybrid trials with external controls, explicitly addressing the tradeoff between causal assumptions and efficiency bounds via semiparametric methods.",{"name":83,"@type":74,"acceptedAnswer":84},"What role do active feature acquisition methods play?",{"text":85,"@type":77},"In precision medicine contexts with costly features, the dissertation extends active feature acquisition to gather only the most relevant information, and further generalizes it to sequential acquisition using an acquisition-conditioned oracle.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"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":107,"slug":139},19,"General","general"]