BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Valid confidence intervals post-model-selection
DTSTART:20231016T073000Z
DTEND:20231016T093000Z
DTSTAMP:20260423T143000Z
UID:indico-event-10709@indico.math.cnrs.fr
DESCRIPTION:Speakers: François Bachoc (IMT / Paul Sabatier)\n\nIn this ta
 lk\, I will first introduce the post-model-selection inference setting\, t
 hat has recently been subject to intensive investigation. In the case of G
 aussian linear regression\, I will review the post-model-selection confide
 nce intervals suggested by Berk et al (2013). These intervals are meant to
  cover model-dependent regression coefficients\, that depend on the select
 ed set of variables. I will present some personal contributions on an adap
 tation of these confidence intervals to the case where the targets of infe
 rence are linear predictors. Then\, I will present an extension of these c
 onfidence intervals to non-Gaussian and non-linear settings. The suggested
  more general intervals will be supported by asymptotic results and numeri
 cal comparisons with other intervals recently suggested in the literature.
 Accès en ligne  https://cnrs.zoom.us/j/98322892547?pwd=d1FpVGwxZUNOVEJRd
 mo0a2NwREZvUT09 \n\nhttps://indico.math.cnrs.fr/event/10709/
LOCATION:salle Fokko du Cloux (1er étage bât Braconnier\, site de la Dou
 a)
URL:https://indico.math.cnrs.fr/event/10709/
END:VEVENT
END:VCALENDAR