{"id":{"repo_id":"buffalo","oai_identifier":"oai:ubir.buffalo.edu:10477/86803"},"canonical_url":"https://search.dev.ndltd.org/etd/buffalo/oai:ubir.buffalo.edu:10477/86803","repository":{"repo_id":"buffalo","name":"Buffalo","base_url":"https://ubir.buffalo.edu/oai/request"},"display":{"title":"A Nuisance-Free Inference Procedure Accounting for the Unknown Missingness Mechanism","abstract":"Ph.D.","abstract_html":"Ph.D.","abstract_has_math":false,"creators":["Chen, Chi"],"institution":"State University of New York at Buffalo","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Zhao, Jiwei","Biostatistics"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-02-25T23:22:55Z","date_published":"2025-02-25T23:22:55Z","updated_at":"2026-07-27T19:05:37Z","subjects":["biostatistics"],"languages":["eng"],"rights":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/10477/86803","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Zhao, Jiwei","Biostatistics"]},{"key":"dc:creator","label":"Author","values":["Chen, Chi"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025-02-25T23:22:55Z","2020","2020-07-08 19:47:42"]},{"key":"dc:publisher","label":"Institution","values":["State University of New York at Buffalo"]},{"key":"dc:type","label":"Dc Type","values":["Text","Dissertation"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["biostatistics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/10477/86803"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Ph.D.","This thesis is motivated by the difficulty of handling missing data in patient-reported outcomes (PRO) and electronic health records (EHR), where the missingness has been well documented as missing not at random (MNAR), or statistically termed as nonignorable. One key challenge in these situations is that, a correct statistical model for the missingness mechanism is extremely difficult, if not infeasible, to specify. To circumvent, we will generally regard the whole problem as a semiparametric model where, other than the parameter of interest, this annoying missingness mechanism is treated as a nuisance nonparametric component. The parameter of interest could come from a commonly-used regression model such as linear regression or logistic regression, depending on the interpretability of the scientific problem. The key idea of this dissertation is to use the conditional likelihoods so that the nonparametric mechanism is completely canceled out through the conditioning arguments. We apply our methodology to a children’s mental health study where the response variable is a PRO and the parameter of interest is from a logistic regression model. We also apply our methodology to the study of the Medical Information Mart for Intensive Care III (MIMIC-III), where the parameter of interest is from a linear regression model. Especially in this latter application, due to the vast amount of available explanatory variables in the EHR system, we also consider the problem of how to statistically solve the high dimensionality issue which includes the variable selection and the post-selection inference. We also conduct comprehensive simulation studies to demonstrate the finite sample performance of our proposed methodology as well as its comparison to some currently existing methods, when the data generation process is known.","**To request an accessible version of the file(s) associated with this item, contact library@buffalo.edu. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.**"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["A Nuisance-Free Inference Procedure Accounting for the Unknown Missingness Mechanism"]}]}],"canonical_facts":{"dc:contributor":["Zhao, Jiwei","Biostatistics"],"dc:creator":["Chen, Chi"],"dc:date":["2025-02-25T23:22:55Z","2020","2020-07-08 19:47:42"],"dc:description":["Ph.D.","This thesis is motivated by the difficulty of handling missing data in patient-reported outcomes (PRO) and electronic health records (EHR), where the missingness has been well documented as missing not at random (MNAR), or statistically termed as nonignorable. One key challenge in these situations is that, a correct statistical model for the missingness mechanism is extremely difficult, if not infeasible, to specify. To circumvent, we will generally regard the whole problem as a semiparametric model where, other than the parameter of interest, this annoying missingness mechanism is treated as a nuisance nonparametric component. The parameter of interest could come from a commonly-used regression model such as linear regression or logistic regression, depending on the interpretability of the scientific problem. The key idea of this dissertation is to use the conditional likelihoods so that the nonparametric mechanism is completely canceled out through the conditioning arguments. We apply our methodology to a children’s mental health study where the response variable is a PRO and the parameter of interest is from a logistic regression model. We also apply our methodology to the study of the Medical Information Mart for Intensive Care III (MIMIC-III), where the parameter of interest is from a linear regression model. Especially in this latter application, due to the vast amount of available explanatory variables in the EHR system, we also consider the problem of how to statistically solve the high dimensionality issue which includes the variable selection and the post-selection inference. We also conduct comprehensive simulation studies to demonstrate the finite sample performance of our proposed methodology as well as its comparison to some currently existing methods, when the data generation process is known.","**To request an accessible version of the file(s) associated with this item, contact library@buffalo.edu. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.**"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/10477/86803"],"dc:language":["eng"],"dc:publisher":["State University of New York at Buffalo"],"dc:rights":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."],"dc:subject":["biostatistics"],"dc:title":["A Nuisance-Free Inference Procedure Accounting for the Unknown Missingness Mechanism"],"dc:type":["Text","Dissertation"]},"updated_at":"2026-07-27T19:05:37Z"}