BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Daniela Witten (University of Washington) DTSTART:20201106T160500Z DTEND:20201106T170500Z DTSTAMP:20260423T005728Z UID:sss/12 DESCRIPTION:Title: Val id hypothesis testing after hierarchical clustering\nby Daniela Witten (University of Washington) as part of Stochastics and Statistics Seminar Series\n\n\nAbstract\nAs datasets continue to grow in size\, in many setti ngs the focus of data collection has shifted away from testing pre-specifi ed hypotheses\, and towards hypothesis generation. Researchers are often i nterested in performing an exploratory data analysis in order to generate hypotheses\, and then testing those hypotheses on the same data\; I will r efer to this as ‘double dipping’. Unfortunately\, double dipping can l ead to highly-inflated Type 1 errors. In this talk\, I will consider the s pecial case of hierarchical clustering. First\, I will show that sample– splitting does not solve the ‘double dipping’ problem for clustering. Then\, I will propose a test for a difference in means between estimated c lusters that accounts for the cluster estimation process\, using a selecti ve inference framework. I will also show an application of this approach t o single-cell RNA-sequencing data. This is joint work with Lucy Gao (Unive rsity of Waterloo) and Jacob Bien (University of Southern California).\n LOCATION:https://researchseminars.org/talk/sss/12/ END:VEVENT END:VCALENDAR