{"id":3999,"date":"2023-05-12T19:08:18","date_gmt":"2023-05-12T19:08:18","guid":{"rendered":"https:\/\/www.stat.matf.bg.ac.rs\/?p=3999"},"modified":"2023-05-16T10:38:18","modified_gmt":"2023-05-16T10:38:18","slug":"%d1%81%d0%b5%d0%bc%d0%b8%d0%bd%d0%b0%d1%80-%d0%ba%d0%b0%d1%82%d0%b5%d0%b4%d1%80%d0%b5-%d0%b7%d0%b0-%d0%b2%d0%b5%d1%80%d0%be%d0%b2%d0%b0%d1%82%d0%bd%d0%be%d1%9b%d1%83","status":"publish","type":"post","link":"https:\/\/www.stat.matf.bg.ac.rs\/en\/%d1%81%d0%b5%d0%bc%d0%b8%d0%bd%d0%b0%d1%80-%d0%ba%d0%b0%d1%82%d0%b5%d0%b4%d1%80%d0%b5-%d0%b7%d0%b0-%d0%b2%d0%b5%d1%80%d0%be%d0%b2%d0%b0%d1%82%d0%bd%d0%be%d1%9b%d1%83\/","title":{"rendered":"Seminar of the Department of Probability and Statistics"},"content":{"rendered":"
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\n

\u0423 \u043f\u0435\u0442\u0430\u043a, 19 \u043c\u0430\u0458\u0430 2023. \u0443 15 \u0447\u0430\u0441\u043e\u0432\u0430, \u0443 \u0441\u0430\u043b\u0438 706, \u0431\u0438\u045b\u0435 \u043e\u0434\u0440\u0436\u0430\u043d\u043e \u043f\u0440\u0435\u0434\u0430\u0432\u0430\u045a\u0435: \u0434\u0440 \u0410\u043d\u0438\u0446\u0430 \u041a\u043e\u0441\u0442\u0438\u045b (LSE – \u041b\u043e\u043d\u0434\u043e\u043d\u0441\u043a\u0430 \u0448\u043a\u043e\u043b\u0430 \u0435\u043a\u043e\u043d\u043e\u043c\u0438\u0458\u0435)<\/p>\n

CHANGE-POINT IDEAS IN MULTIPLE TESTING<\/strong><\/p>\n

\u0420\u0435\u0437\u0438\u043c\u0435: Estimating the proportion of false null hypotheses among independently tested hypotheses is a crucial problem in the multiple testing literature. By integrating a proportion estimator, most multiple testing methods that control the false discovery rate (FDR) can be made adaptive, leading to increased power while still maintaining control over the FDR. Although many proportion estimators have been proposed in the literature, the approach by Schweder and Spjotvoll, more commonly known as Storey\u2019s estimator, is the most widely used due to its simplicity. We present a novel approach that tunes Storey’s estimator using ideas from the change-point literature. Our approach aims to identify the change-point in the slope of the p-value plot to distinguish between true and false null hypotheses. The proposed method is fast, simple, and improves on existing tuning parameters. The theoretical results for our estimator are based on strong limit theorems for quantile processes. The simulation study shows that our method outperforms existing methods, particularly in sparse cases with a small false null proportion. Additionally, we discuss the performance of our method when the p-values are dependent or superuniform, which typically causes Storey’s estimator to fail.<\/p>\n

\u0421\u0435\u043c\u0438\u043d\u0430\u0440\u0443 \u0458\u0435 \u043c\u043e\u0433\u0443\u045b\u0435 \u043f\u0440\u0438\u0441\u0443\u0441\u0442\u0432\u043e\u0432\u0430\u0442\u0438 \u0438 \u043e\u043d\u043b\u0430\u0458\u043d.
\u041b\u0438\u043d\u043a \u0437\u0430 \u043f\u0440\u0438\u0441\u0442\u0443\u043f \u0458\u0435\u00a0Join Zoom Meeting<\/a>
Meeting ID: 965 1362 4215 Passcode: 704826<\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>","protected":false},"excerpt":{"rendered":"

\u0423 \u043f\u0435\u0442\u0430\u043a, 19 \u043c\u0430\u0458\u0430 2023. \u0443 15 \u0447\u0430\u0441\u043e\u0432\u0430, \u0443 \u0441\u0430\u043b\u0438 706, \u0431\u0438\u045b\u0435 \u043e\u0434\u0440\u0436\u0430\u043d\u043e \u043f\u0440\u0435\u0434\u0430\u0432\u0430\u045a\u0435: \u0434\u0440 \u0410\u043d\u0438\u0446\u0430 \u041a\u043e\u0441\u0442\u0438\u045b (LSE – \u041b\u043e\u043d\u0434\u043e\u043d\u0441\u043a\u0430 \u0448\u043a\u043e\u043b\u0430 \u0435\u043a\u043e\u043d\u043e\u043c\u0438\u0458\u0435) CHANGE-POINT IDEAS IN MULTIPLE TESTING \u0420\u0435\u0437\u0438\u043c\u0435: Estimating the proportion of false null hypotheses among independently tested hypotheses is a crucial problem in the multiple testing literature. By integrating a proportion estimator, most multiple […]<\/p>\n","protected":false},"author":4,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-3999","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"blocksy_meta":{"styles_descriptor":{"styles":{"desktop":"","tablet":"","mobile":""},"google_fonts":[],"version":8}},"_links":{"self":[{"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/posts\/3999","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/comments?post=3999"}],"version-history":[{"count":12,"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/posts\/3999\/revisions"}],"predecessor-version":[{"id":4187,"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/posts\/3999\/revisions\/4187"}],"wp:attachment":[{"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/media?parent=3999"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/categories?post=3999"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.stat.matf.bg.ac.rs\/en\/wp-json\/wp\/v2\/tags?post=3999"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}