BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Iowa State University CALS LAS Web Team//sites.iastate.edu//EN
BEGIN:VEVENT
UID:20211101T160000-905-www.stat.iastate.edu
DTSTART:20211101T160000Z
SEQUENCE:0
TRANSP:OPAQUE
DTEND:20211101T170000Z
LOCATION:Zoom: https://iastate.zoom.us/j/96431743984?
SUMMARY:Daniela Witten\, University of Washington: Selective inference for 
 trees
CLASS:PUBLIC
DESCRIPTION:Abstract:&nbsp\;As datasets grow in size\, the focus of data co
 llection has increasingly shifted away from testing pre-specified hypothes
 es\, and towards hypothesis generation. Researchers are often interested i
 n performing an exploratory data analysis to generate hypotheses\, and the
 n testing those hypotheses on the same data. Unfortunately\, this type of 
 'double dipping' can lead to&nbsp\;highly-inflated Type 1 errors. In this 
 talk\, I will consider double-dipping&nbsp\;on&nbsp\;trees. First\, I will
  focus on&nbsp\;trees&nbsp\;generated via hierarchical clustering\,&nbsp\;
 and will consider testing the null hypothesis of equality of cluster means
 .\n\nMore information at: https://www.stat.iastate.edu/event/2021/daniela-
 witten-university-washington-selective-inference-trees\n\nZoom: https://ia
 state.zoom.us/j/96431743984?
DTSTAMP:20261015T141549Z
END:VEVENT
END:VCALENDAR