BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Bayesian Superiority in On/Off analysis
DTSTART;VALUE=DATE-TIME:20261030T135500Z
DTEND;VALUE=DATE-TIME:20261030T140000Z
DTSTAMP;VALUE=DATE-TIME:20261107T204323Z
UID:indico-contribution-6469@cern.ch
DESCRIPTION:Speakers: Petr Satunin (INR RAS)\nWe present a detailed compar
 ison of Bayesian criteria with three non-informative priors - flat\, Jeffr
 eys\, and scale-invariant - for testing a signal against an unknown backgr
 ound and compare them with the classical frequentist Li-Ma approach in the
  On/Off problem. We perform Monte Carlo simulations for various background
  levels and evaluate the Li-Ma and Bayesian criteria by their Type I error
  rates. We then simulate a nonzero signal and compare the criteria in term
 s of Type II error rates. We find that both the Li-Ma criterion and the Ba
 yesian criterion with the Jeffreys prior control Type I error well\, exclu
 ding the case of small background b\n\nhttps://indico.particle.mephi.ru/ev
 ent/626/contributions/6469/
LOCATION: Hall
URL:https://indico.particle.mephi.ru/event/626/contributions/6469/
END:VEVENT
END:VCALENDAR