When: Friday, February 13, 3:00 PM
\nWhere: Tyler 055<\/p>\n
Abstract<\/strong> In this talk, I introduce a nonlinear DiD framework that addresses three key limitations of classical approaches: it accommodates nonlinear trends, is invariant to outcome transformations, and captures effects across the outcome distribution\u2014not just the mean.<\/p>\n Our method features a novel identification strategy that allows for high-dimensional, non-monotonic unmeasured confounding, and employs modern machine learning and semiparametric tools to construct efficient, debiased estimators. I will present new theoretical results on identification and inference, and illustrate the method\u2019s performance through simulations and a case study on the political impacts of mass shootings in U.S. counties.<\/p>\n This work is part of a broader research agenda at the intersection of causal inference, temporal data, and real-world decision-making. I\u2019ll conclude by outlining future directions in panel and trajectory-based causal methods, with applications in public health, medicine, and the social sciences.<\/p>\n Bio<\/strong> When: Friday, February 13, 3:00 PM Where: Tyler 055 Abstract Difference-in-differences (DiD) is a foundational tool for causal inference in panel data, widely used in policy, economics, and health research. Its appeal lies in its intuitive design and robustness to time-invariant unmeasured confounding. However, standard DiD relies on strong assumptions\u2014particularly parallel trends\u2014that are often violated […]<\/p>\n","protected":false},"author":3300,"featured_media":18180,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":"","_links_to":"","_links_to_target":""},"categories":[34,85],"tags":[],"class_list":["post-18177","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news","category-seminars"],"acf":[],"_links":{"self":[{"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/posts\/18177","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/users\/3300"}],"replies":[{"embeddable":true,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/comments?post=18177"}],"version-history":[{"count":2,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/posts\/18177\/revisions"}],"predecessor-version":[{"id":18179,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/posts\/18177\/revisions\/18179"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/media\/18180"}],"wp:attachment":[{"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/media?parent=18177"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/categories?post=18177"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/web.uri.edu\/cs\/wp-json\/wp\/v2\/tags?post=18177"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\nDifference-in-differences (DiD) is a foundational tool for causal inference in panel data, widely used in policy, economics, and health research. Its appeal lies in its intuitive design and robustness to time-invariant unmeasured confounding. However, standard DiD relies on strong assumptions\u2014particularly parallel trends\u2014that are often violated in practice.<\/p>\n
\nJinghao Sun is a Postdoctoral Researcher at the University of Pennsylvania\u2019s Center for Causal Inference. His work focuses on causal inference and randomized trial methods for complex temporal data (panel, survival, and functional data), motivated by problems in public health, medicine, and policy. He earned his PhD in Biostatistics from Yale University in 2023.<\/p>\n","protected":false},"excerpt":{"rendered":"