BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/568
DTSTAMP:20230914T125929Z
SUMMARY:Information Geometry of the Robust Maximum Mean Power Likelihood Es
timation
DESCRIPTION:Speaker: Rajesh Sundaresan (Indian Institute of Science\nDepart
ment of Electrical Communication Engineering\nBangalore 560012)\n\nAbstrac
t: \nAbstract: The maximum "mean power" likelihood estimation procedure (p
roposed by [Basu et al.\, Biometrika\, 1998]) is a robust variant of the m
aximum likelihood estimation procedure. Given iid samples coming from an u
nknown distribution\, a member of a given parametric family of distributio
ns\, the mean power likelihood estimation procedure is to find the paramet
er of the distribution that is closest\, in a particular sense\, to the em
pirical distribution of the samples. One may view the resulting distributi
on as a "projection" of the empirical distribution on the parametric famil
y. In this talk\, I will highlight the geometry associated with this proje
ction. I will also discuss a simplified computation procedure\, one that i
s suggested by the geometric view point\, when the estimation is of the pa
rameter of a power-law family (this is joint work with M. Ashok Kumar). ar
Xiv:1410.5550\n\nBiography: Rajesh Sundaresan is an associate professor at
the ECE department of IISc Bangalore.\n \n
URL:https://www.tcs.tifr.res.in/web/events/568
DTSTART;TZID=Asia/Kolkata:20150120T134500
DTEND;TZID=Asia/Kolkata:20150120T144500
LOCATION:AG-69
END:VEVENT
END:VCALENDAR