BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Cavendish HEP Seminars
SUMMARY:Custom Orthogonal Weight functions (COWs) - an imp
roved event weighting procedure for removing backg
round from signal. - Matt Kenzie (University of Wa
rwick)
DTSTART;TZID=Europe/London:20230124T110000
DTEND;TZID=Europe/London:20230124T120000
UID:TALK192911AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/192911
DESCRIPTION:A common problem in data analysis is the separatio
n of signal and background. This talk is based on
https://arxiv.org/abs/2112.04574 where we revisit
and generalise the so-called sWeights method\, whi
ch allows one to calculate an empirical estimate o
f the signal density of a control variable using a
fit of a mixed signal and background model to a d
iscriminating variable. We show that sWeights are
a special case of a larger class of\nCustom Orthog
onal Weight functions (COWs)\, which can be applie
d to a more general class of problems in which the
discriminating and control variables are not nece
ssarily independent and still achieve close to opt
imal performance. We also investigate the properti
es of parameters estimated from fits of statistica
l models to sWeighted data and provide closed form
ulas for the asymptotic covariance matrix of the f
itted parameters. To illustrate our findings\, we
discuss several practical applications of these te
chniques.
LOCATION:Ryle Seminar Room
CONTACT:William Fawcett
END:VEVENT
END:VCALENDAR