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:Statistical Laboratory Graduate Seminars
SUMMARY:An overview of shape-constrained estimation proble
ms - Yining Chen (University of Cambridge)
DTSTART;TZID=Europe/London:20120312T140000
DTEND;TZID=Europe/London:20120312T150000
UID:TALK35861AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/35861
DESCRIPTION:Shape-constrained density estimation has received
a great deal of interest recently. The allure is t
he prospect of obtaining fully automatic nonparame
tric estimators with no tuning parameters. The gen
eral idea dates back to Grenander (1956)\, who der
ived the maximum likelihood estimator of a decreas
ing density on [0\,∞). Some other popular shape-co
nstraints include convex and log-concave.\n\nIn th
is talk\, I will give a brief overview of the area
\, focusing particularly on the log-concave constr
aint (i.e. the logarithm of the density function i
s concave). I will also mention nice applications
of this technique in regression problems and time
series analysis. Essential background of nonparame
tric statistics will also be covered with R demons
trations.
LOCATION:CMS\, MR12
CONTACT:Elena Yudovina
END:VEVENT
END:VCALENDAR